Livestock Research for Rural Development 17 (1) 2005 Guidelines to authors LRRD News

Citation of this paper

Typology of dairy cattle farming systems in the Gharb irrigated perimeter, Morocco

M T Sraïri and N Kiade

Animal Production Department, Hassan II Agronomy and Veterinary Institute
P.O. Box 6202, Rabat-Institutes, 10101, Rabat, Morocco
tsrairi@yahoo.fr


Abstract

The characterisation of dairy farming systems in the Gharb irrigated perimeter, located in the north-western plains of Morocco, was achieved through regular enquiries of seventy farms, during 2000/2001 agricultural campaign. Main objective of this study was to describe diversity in cattle farming practices, in a context of limited references, in order to plan further development efforts, to enhance cattle productivity.

Results show a broad variety of farmers' strategies, particularly in economic and feeding efficiency, in types of breeding practices (intensification vs. extensive farming) and in functions devoted to cattle (milk and/or meat). In order to avoid a typology that would only make difference between big farms and smallholders units, two state farms were put away from multivariate analysis. Five groups of cattlemen were then distinguished based upon multivariate statistical analysis (principal components and cluster). The first one was represented by 16 extensive dairy farms with limited use in concentrates, and as a result low milk yield (1998 kg/cow). The second group relied much heavily on forages that made up for 5.65-times concentrates energy intake by cows. Consequently, group 2 had better economic profitability (324.7 vs. -43.2 US $ per cow in groups 2 and 1 respectively), with reduced costs of production. Group 3 illustrated the case of a more intensive feeding strategy which resulted in a higher milk yield per cow (3223 kg). Group 4 was identified as concentrate wasters who could not cope with limited forages and did not monitor enough cows feeding formulas. Finally, group 5 was made of 5 big units (average arable land = 37.0 ha), who did not invest enough on dairy production. They were more interested in other lucrative cash crops (sugar beet, tomatoes and potatoes). Consequently, in those units, cattle seem to have other functions linked to social prestige, job creation, and even providing milk to farmers' families.

It was concluded, that this diversity in dairy systems should be taken into account for development purposes. Specific measures need to be applied to each one of the livestock systems identified, mainly with the extension of feed formulas using locally available by-products. Another axis of intervention would be to promote meat production practices (feedstuffs and breeds), as dairy specialisation is far from reached in all farms.

Key-words: Dairy cattle, economic profitability, farming systems, husbandry practices, morocco, typology.


Introduction

Since the early 60's, ambitious plans of dams building have led to the irrigation of more than 800,000 ha in Morocco. In those areas, priority has been given to high output production, such as intensified dairy cattle farming, in order to fulfill increasing demand for animal proteins, in a context of rapid demographic expansion. With more than 10% of the country's milk production and with a convenient geographical location which provides enough rainfall in comparison to drier regions of the south and the east parts of the country, Gharb irrigated perimeter was very soon identified as the most suited area for intensive dairy breeding in Morocco (Projet Sebou 1961). A sharp decrease in dairy production has however been reported in recent years at the regional level when compared to other irrigated zones (Sraïri and Ilham 2000). On the other hand, very few aspects of dairy farming systems have been studied in Morocco, with an overview approach, as systemic studies were applied successfully to assess animal production projects in other regions of developing countries (Landais 1983; Schiere and de Wit 1993). Moreover, Roeleveld and Van den Broek (1996) emphasised the important role for a livestock performance diagnosis, as a preliminary step to any promoting project, especially in developing countries. Systemic studies based on animal husbandry practices diagnosis can also be an efficient way to examine the relations of farms to space utilisation and time, which could be of significant interest for understanding farms' evolution (Girard et al 2001). Hence, the lack of information on dairy performances and on farmers' practices can only hinder relevant efforts to enhance milk yield at farm-scale. The existing references focused solely on high-input farms (Lakhdissi et al 1988; Sraïri and Kessab 1998), practically ignoring the situation of smallholders, which represent 83% of the global number of dairy farms, as they have more than 60% of the country's cows.

The present article aims to analyse farmers' strategies and management practices to produce milk, in order to elaborate a typology of dairy farms in the Gharb perimeter. Such a typology would be a practical tool for further development effort in the dairy cattle industry, as it would allow for the implementation of adapted measures to targeted groups of farmers (Sraïri 2001).


Material and methods

Data collection

A benchmark survey of dairy cattle breeders was conducted in the Gharb irrigated perimeter from September 2000 to July 2001. A total of 70 cattle owners were selected with the approval of breeders' associations, and the only condition considered was milk production and marketing, regardless to herd structure or farm level of intensification. Type proportions within the final typology may not reflect the reality. Indeed, a large number of big herd breeders was deliberately chosen to involve a significant sample of each type in the study. It was apparent in the field that large herd owners were fewer, in proportion, than what is described in this article. Accordingly, the results of the present study did not respect the actual distribution of the various types of cattle farmers, especially in relation with their structural characteristics. Actually, recent data provided by general census of cattle in Morocco, and especially in Gharb perimeter indicate that 77 % of cattle are in smallholder units with less than 3 cows on less than 5 ha of arable land (MADRPM 2002). In our study sample, there was an average of 7.5 cows, and 14.7 ha of arable land per farm, due to the selection of big units, of which two were state farms. Herders were frequently interviewed and their herd performances were followed-up. Data about structural parameters (arable land, number of cows, irrigation possibilities...), dairy cattle management (feeding, milking frequency, veterinary treatments and reproduction) and economic results were collected. A ten-page data form was filled for each dairy cows unit. It was completed by three specific visits to each farm. These visits were separated by approximately 100 days in order to assess milk production performances and to determine costs and income generated from cows during a whole agricultural campaign. The observations were divided into four main parts. First one covered cattle farms structural parameters, while the second dealt with dairy cattle feeding (purchases of concentrates, utilisation of forages...). The third part investigated into cows' reproduction, and the last into economic results of dairy herds, after the analysis of total inputs (forages, concentrate feeds, veterinary products...) and global animal products sails (milk, cattle and even manure).

Data analysis

A typology of cattle farming systems was set up. It took into account different elements of a farming system, i.e., the farmer (income, patrimony and history), the herd (composition and technical and economic performance), and the resources involved in dairy cattle farming (Gibon et al 1999). Statistical analyses were run with the multivariate analysis software package SAS (SAS 1998). At a first step, a total of 12 quantitative variables were identified to describe dairy farms and their activities during 2000/2001agricultural campaign. A principal components analysis (PCA) was used to detect main variables that characterise farms sample (Procedures PROC FACTOR and PROC PRINCOMP). At this point, it appeared that two farms had structural parameters far from mean value and they resulted in structural plan (axis 1 and 2) with reduced interest as they discriminated between big farms and smallholders' units. Therefore, we decided to put away those two farms and we ran a new PCA. It was obvious that the main variables describing sample of dairy farms had no more relation to structural parameters (i.e. number of cows and arable land), but were determined by feeding management of the herds, cattle milk yield and economic efficiency per cow, combined to the weight of cattle sales compared to milk sales. A group of 10 variables were finally used to achieve the characterisation of breeding practices diversity (Table 1). A cluster analysis was then run to elaborate a final distribution of farms into homogenous groups (PROC CLUSTER). A typology made of five groups was retained to synthesise the diversity of cattle farming practices in the Gharb perimeter.

Table 1. Variables used in principal components analysis to characterise trends in dairying in Gharb irrigated perimeter, Morocco 

Variables

Symbol

Cattle Load, ha of forages/cow

CLO

Variation in Cattle Number, Cattle Unit

VCN

Milk Yield per Cow per year

MYC

Concentrates Energy per Cow per year, Mcal

CEC

Concentrates Energy per kg of Milk, Mcal

CEK

Forages to Concentrates Ratio, %

FCR

Cattle / Milk in Global Sales, %

CMG

Feed / Total Costs, %

FTC

Production Cost per kg of Milk, US $

PCM

Gross Margin per Cow, US $

GMC


Results

Herd structure and production efficiency

Average parameters describing sampled dairy farms are summarised in Table 2.

Table 2. Dairy farms characteristics in the Gharb perimeter

Parameters

Minimum

Average
± standard  deviation

Maximum

Arable land, ha

       0.5

        14.80 ± 46.8

         338

Forage area, ha

       0.1

          2.71 ± 3.35

           20.8

Cattle load, ha of forages per cow

       0.0

          0.48 ± 0.42 

             3.4

Number of cows

       1

          7.5 ± 13.5

         106

Total milk production per herd, kg

   801

  22014 ± 74460

   699368

Milk yield per cow, kg

   727

    2588 ± 1121

       5909

Concentrates*/cow/year

   271.2

    2019 ± 1292

       6730.4

Concentrates*/kg of milk

       0.15

          1.00 ± 0.64

             3.91

Concentrates/Total energy supply, %

     12.2

        46.3 ± 40.2

           83.3

Cattle/milk in gross product, %

       8.3

        73.0 ± 81.3

           76.9

Feed/total costs, %

     44.1

        77.8 ± 14.4

           98.9

Production Cost per kg of Milk, US $

       0.16

          0.34 ± 0.19

             1.12

Gross margin per cow, US $

 - 965.2

      172.7 ± 258.4

       1252.2

*Amounts of concentrates expressed in Mcal of Net Energy

Because of the presence in the studied sample of farms with structural parameters out of range than "normal" farms, a huge dispersion of these parameters was observed (i.e standard deviation superior to mean values). For example, there were 14.8 ± 46.8 ha of arable land per farm, with more than 80% of the farms using less than 22% of total land. Forage reserved areas represented 18% of total land and were mainly made of bersim (Trifolium alexandrinum) (i.e 60% of total forages land), followed by alfalfa and maize.

Genetic structure was dominated by Holstein Friesian cows (58% of total cattle). They were followed by crossbred local x Holstein Friesian cows (39%) and ultimately by local genotype cows (3%). The analysis of cattle feeding showed that concentrates represented 46.3% of total energy ingested by cows. The important use of concentrates, and the diversity of feeding formulas, often resulted in unbalanced rations, as concentrates contributed to maintenance requirements in many occasions, in a situation where forages were lacking.

Reproduction performances, calculated for 72 cows, were rather unsatisfactory, with an average interval between calving of 435 days. Dairy farms economic results were characterised by a wide variation, from positive results to deficit. When average profitability generated by one cow was 131 US $, it varied widely from -965 to 1252 US$. On average, feeding costs represented more than 77% of total inputs, showing the importance of feeding practices not only on global economic results but on the whole strategies adopted by farmers for production purposes.

Multivariate analysis

The main goal of the multivariate statistical analysis was to emphasise the types of links between descriptive variables that characterise farms and to create homogenous groups of dairy units in relation to predominant variables. In the second and final PCA, three first axes accounted for 72.6% of total variation of the farms sample (Table 3).

Table 3. Results of the Principal Components analysis – Axis definition

Axis

Axis definition

Proportion, %

Cumulative variation, %

Variables

Correlation to axis

1

Animal load

- 0.63

 

 

 

Concentrates / kg of milk

  0.86

 

 

 

Forages / concentrates ratio

- 0.76

31.6

31.6

 

Production Cost / kg of milk, US $

- 0.66

 

 

2

Milk yield / cow / year

  0.86

 

 

 

Concentrates / cow / year

  0.79

21.9

53.5

3

Cattle / milk in gross product

  0.84

 

 

 

Variations in cattle number

- 0.64

19.1

72.6

First axis represented 31.6% of total variation. It was interpreted as an axis of feeding and economic efficiency. It was highly correlated to variables such as concentrates use per kg of milk and production cost per kg of milk. Thus, it differentiated between efficient herds in converting concentrates in milk, with adequate supply of forages in global energy intake, to herds with opposite characteristics. As a matter of fact, this axis took also into account variables such as animal load (ha of forages per cow) and forages contribution to energy intake by cows compared to concentrates.

The second axis explained 21.9% of total variation and was positively correlated to variables "amount of concentrates used annually per cow" and "milk yield achieved per cow". It was defined as an axis of intensification in dairying, as it opposed farms with reduced amounts of concentrates and milk yields per cow to farms with more intensive concentrates use that results in higher milk production per cow (Figure 1).

Figure 1. Projections of variables on PCA primary plan defined by axes 1 and 2

The third axis represented 19.1% of total variation and was closely correlated to the variation of cattle numbers in herds during the agricultural campaign 2000/2001 and to the ratio cattle sales/milk sales. We interpreted this axis as an axis of specialisation in dairy production as it opposed specialised dairy farms with few cattle sales in comparison to milk sales, and with steady herd composition, to farms with active movements in the composition of the herd and important cattle/milk ratio in global sales.

A cluster analysis was then conducted using results from the second principal components analysis. Five groups were finally selected (Figure 2).

G1: Group 1, G2: Group 2, G3: Group 3, G4: Group 4, G5: Group 5


Figure 2.
Farms group representation in the PCA by axis 1 and 2

Group 1 can be considered as "typical extensive cattle farming". It gathers 16 farms with reduced milk yield (1998 kg per cow), due to insufficient use of concentrates per cow. Even if this kind of practice results in a low cost to produce a kg of milk, gross margin per cow is negative (- 43.2 US $), because maintenance requirements are not compensated enough by milk incomes. This makes us assume that in this kind of system cattle surely have non market functions: providing family members with dairy products and participating in soil fertility preservation by the production of dung.

Group 2 is representative of "successful extensive dairying in smallholders units, under Moroccan conditions". It is constituted by 10 farms, which are all profitable, with an average gross margin per cow of 324.7 US $. The most obvious characteristic in these farms is the high level of forages in global intake of energy, as they represented 5.65-times energy from concentrates. This was achieved due to less cattle load than in the first group (1.2 versus 2.1 ha of forages/cow, respectively in groups 1 and 2).

Group 3 is an illustration of a "beginning of intensification in dairy cattle feeding practice". With 29 farms, it seemed very diverse in its composition, due to a large dispersion of units in it. A global majority of farms showed profitable dairy activity (322.3 US $ of gross margin per cow). This was a result of a beginning of intensification, as energy from concentrates per cow reaches an average of 2604 Mcal vs. 2019 Mcal for the whole sample of studied farms. This intensification had evident impacts on milk yield and on economic profitability.

The 8 farms in group 4, had a main characteristic of "concentrate wastage". Even with an energy intake from concentrates of 2620 Mcal per cow, average milk yield remains weak (1486 kg). Production cost for one milk kg is therefore very high (0.75 US $), and this situation is not equilibrated by cattle sales. Consequently this group illustrates the case of smallholders units, which do not manage to cope with reduced forages area and which waste concentrates in unbalanced feed formulas.

Group 5, with 5 farms, is made of "average milk yield and economic results per cow lying on farms with important production means". Their main characteristics are close to mean values calculated for all studied samples (2486 kg of milk yield and 110 US $ of gross margin per cow). However, in these farms it seems that average structural parameters (37.0 ha of arable land and 19 cows) are much higher than global means for those parameters (respectively 7.3 ha and 5.5 cows). Obviously, these 5 farms are bigger than the average Gharb farms, but it seems that dairy practices adopted (feeding mainly) are very extensive (less than 1498 Mcal of concentrates per cow). Thus, a real dairy potential is being wasted on these farms.

Finally, last two farms, taken away after the first PCA, are belonging to a state society (Society of Agricultural Development, SODEA). Those farms illustrate a peak of intensification in dairying at the regional scale, with an average 5909 kg of milk yield, 361.2 US $ of gross margin, and an intake of 6273 Mcal of concentrates per cow.


Discussion

Farming systems are usually defined through the interactions of breeders, herds and environmental conditions and resources (Lhoste 1984). The multivariate analysis conducted in this study showed that feeding practices (i.e., resources) data were predominant.

Focus should then be on the multivariate global analysis and the typology results including both technical and economic performances of different herd sizes. In a recent study conducted under Polish conditions, Kaminiecki et al (1999) have reported significant differences in family dairy cattle farming systems. Laval et al (1998) have also conducted a study to assess the diversity of camel farming systems in Rajasthan (India), using multivariate analysis as a tool to describe variations between groups of farms.

Farmers of groups 1, 2, 3 and 4 were the most frequently observed in the sample. Those four groups represented 63 of the 70 farmers sampled. Structural parameters observed (arable land and cow numbers) and dairy yields per cow, clearly illustrate the case of smallholders production, which represents more than 80% of cattle farms in Morocco. Differences between those four classes should not be found in their structural parameters but mainly in the way milk production is organised and as a consequence in their economic results per cow and in the efficiency of concentrates conversion to milk (economic profitability versus deficit, concentrates valorised in milk versus concentrates wasted in cows maintenance requirements). Furthermore, feeding strategies (intensive versus extensive dairying) and milk versus meat production choices by breeders were also predominant aspects to classify farms.

Group 1 illustrates clearly smallholders with insufficient use of energy from concentrates to achieve enough milk yield per cow (1998 kg) (Table 4). Thus, as reported by Wolter (1994), these farmers do not manage to valorise forages and all other inputs they use, because of restricted energy offer to the cows. Group 2 represents another strategy in cows' feeding, as forages account  for 5.65-times the energy value of concentrates. There is a clear choice of satisfying cows requirements mainly on forages with the minimum purchases of concentrates. Group 3 is typically gathering intensive smallholder dairy farms in our study sample. This intensification was a result of more concentrate consumption (2604 versus 2019 Mcal per cow). Thus, economic profitability is better than the average (312.3 versus 172.7 US $ per cow). This beginning of specialisation in milk production vs. meat can be seen through cattle sales, which only represent 36.7% of milk sales.

Table 4. General comparison between dairy farm groups identified by cluster analysis

Group or farm

1

2

3

4

5

SODEA

Number of farms

16

10

29

8

5

2

Arable land, ha

         3.97

         6.65

         5.20

        3.96

      37.00

     338.35

Number of cows

         3.37

         3.17

         5.90

        3.04

      19.00

     106.5

Cattle Number Variation

         0.46

         1.25

        -1.02

       -0.94

      -1.24

        -1.20

Milk yield/cow, kg

  1998

 2326

  3223

  1486

 2486

  5909

Concentrates*/cow

   1154

   847

  2604

  2618

 1498

  6732

Forage/concentrates, %

   289.5

  564.8

    115.5

      57.1

   205.8

      12.0

Cattle/milk sales ratio, %

    36.8

  112.5

      64.7

    145.1

     55.4

      12.0

Cost / kg of Milk, US $

      0.31

      0.27

        0.28

        0.74

       0.27

        0.26

Gross Margin/cow, US $

   - 43.2

  324.7

    312.3

    230.8

   110.2

    361.2

* Amounts of concentrates expressed in Mcal of Net Energy

Group 4 is made of concentrate wasters. This is the case of 8 farms, which do not monitor well the production cost per kg of milk (0.74 US $). This is due to the use of too much concentrates (2618 Mcal per cow) but which are very badly converted into milk (only 1486 kg per cow). This is a direct consequence of lack of extension of feed formulation techniques.

Group 5 is very different from the others, as it is represented by 5 farms which are not "smallholders" structures anymore. However, technical practises adopted do not show any dairy specialisation, as concentrates offered to cows is limited to 1498 Mcal per cow. Actually, it is clear that those farms are neglecting their herds, because of other priorities (industrial and horticultural crops, such as sugar beet, tomatoes, or sunflower). One reason frequently advanced by breeders in this category is reduced cash flow generated by cows in comparison to crops, particularly in a context where raw milk price has not increased since early 90's (Akesbi 1997). However, cattle are kept on farms because of their numerous functions (job opportunities, soil fertility, social prestige, capital stocking...).

Finally, the description of dairy systems existing in Gharb region would be incomplete without big farms, with a deliberate strategy of intensive dairying. This is the case of two SODEA farms taken away from the general PCA adopted in the early step of this study. In fact, both farms are belonging to a state society, which was created in the early 70's to enhance productivity of Moroccan agriculture, on farms formerly owned by foreign settlers. Their structural parameters and their overall results are totally out of scale compared to other situations. These kind of farms, which are heavily advised by engineers and technicians, especially in the field of feed formulation practice, also have substantial gains in feed purchases, as they have access to cheaper offers. Similar breeding characteristics were also identified on other state farms (Sraïri and Kessab 1998).

For development perspectives, it is obvious that research adapted to smallholders is urgently needed. From this point of view on-farm trials should have priority, as the main idea would be to work with farmers to find solutions (Stür et al 2002). Technical advice in the field of feed formulation and forages exploitation and conservation by ensiling means should be effective. As concentrates are widely used, it's their mixing in adapted formulas to variable forages that could provide significant results. On another hand, modern facilities in the field of reproduction (hormone treatments in repeat breeding cows) and AI generalisation can be effective tools to enhance herds breeding value. Prophylactic methods should also be set up for various parasitic diseases, in a context where few herds get preventive treatments. All categories of breeders, with a special reference to smallholders should be handled. And because dairy specialisation is far to be uniform, targeted measures are necessary, to avoid the failure of technology transfers, as it has happened with urea-treatments for straw (Wanapat et al 1998). Therefore, high capital needing techniques should be avoided. Extension of feed formulation techniques would probably have significant results. However, choosing relevant farmers (educated and motivated ones) is a necessary condition to achieve success, whether for high milk yielding herds, or for more meat oriented units. Then, those farmers could show the way to others, as the key idea behind such development projects is not to pretend to teach farmers but to be aware that there is much to learn from them (Dumont 1974).

Dairy cattle farming is still one of the most efficient ways to enhance income in rural areas of Morocco, with a growing demand for animal proteins. In such a case, farming typologies and appropriate development work are compulsory. This should be seriously taken in consideration by development decision-makers.


Conclusions


Acknowledgements

 

The authors are thankful to all the Gharb breeders enquired for their kind participation in the field survey, and particularly Mr Khalid, director of a milk collecting centre, for his timely help. Technical assistance provided by M. Chaaïbi A. from the Gharb Regional Office of Agricultural Production (O.R.M.V.A.G.) and logistical support from the Centrale Laitière Company are duly acknowledged.



References

 

Akesbi N 1997 La question des prix et des subventions au Maroc face aux mutations de la politique agricole. Options Méditerranéennes, Série B, n° 11, Prix et subventions: effets sur les agricultures familiales méditerranéennes. p. 81 - 117.

 

Dumont R 1974. Agronome de la faim. Collection Un homme et son métier. Editions. Robert Laffont. Paris, France, 394 p.

 

Gibon A, Sibbald A R, Flamant J C, Lhoste P, Revilla R, Rubino R and Sorensen J T 1999 Livestock farming systems research in Europe and its potential contribution for managing towards sustainability in livestock farming. Livestock Production Science. 61 121-137.

 

Girard N, Bellon S, Hubert B, Lardon S, Moulin C H and Osty P L 2001. Categorising combinations of farmers land use practices: an approach based on examples of sheep farms in the south of France. Agronomie. 21 435-459.

 

Kamieniecki K, Gnyp J and Trautman J 1999 Technical and economic characteristics of dairy farms in Central-East Poland. Livestock Production Science. 61 301-306

 

Lakhdissi H, Lahlou-Kassi A and Thibier M 1988 Conduite de la reproduction en grands troupeaux laitiers dans les conditions marocaines. I. Influence du programme d'action vétérinaire intégré de reproduction sur les bilans de fertilité. Revue d'Elevage et de Médecine vétérinaire des Pays Tropicaux. 41 293-299

 

Landais E 1983 Analyse des systèmes d'élevage bovin sédentaire du nord de la Côte d'Ivoire. Maisons-Alfort, France. IEMVT, Volume 1 and 2, 789 p.

 

Laval G, Khanna N D and Faye B 1998 A typology of camel farming systems in Bikaner and Jaisalmer districts of Rajasthan, India. Revue d'Elevage et de Médecine vétérinaire des Pays Tropicaux. 51 147-154

 

Lhoste P 1984 Le diagnostic sur le système d'élevage. Cahiers de la Recherche-Développement. Volume 3-4, 84-88.

 

MADRPM 2002 General agricultural census. Cattle. Results for the perimeter of the Gharb (in French). Rabat, Morocco. 154 p.

 

Projet Sebou 1961 Mise en valeur agricole de la Plaine du Gharb. Elevage bovin. Fascicule 4, Livret 2. Rabat, Morocco. 85 p.

 

Roeleveld A C W and Van Den Broek A 1996 Focusing livestock systems research. Royal Tropical Institute, Amsterdam. 151 p.

 

SchiereJ  B and De Wit C T 1993 Feeding standards and feeding systems. Animal Feed Science and Technology. 43 121-134

 

Sraïri M T 2001 Déterminisme et applications de la recherche systémique pour l'étude de l'élevage laitier. Courrier de l'Environnement de l'INRA (France). 42, pp. 29-46.

 

Sraïri M T and Ilham A 2000 L'élevage laitier en zone irriguée et sa sensibilité à l'aléa climatique: cas du Maroc. In: Guessous, F., Rihani, N., Ilham A., (Editors.), Livestock production and climatic uncertainty in the Mediterranean. EAAP Publication 94, pp. 321-327.

 

Sraïri M T and Kessab B 1998 Performances et modalités de production dans six étables laitières intensives au Maroc. INRA Productions Animales. 11 299-304.

 

SAS 1998 User's Guide: Statistics, Version 6. SAS Inst., Iric., Cary, NC. USA.

 

Stür W W, Horne P M, Gabunada Jr F A, Phengsavahn P and Kerridge P C 2002 Forage options for smallholder crop-animal systems in Southeast Asia: working with farmers to find solutions. Agricultural Systems. 71 75-98.

 

Wanapat M, Chenost M, Muñoz F and Kayouli C 1996 Methods for improving the nutritive value of fibrous feed: treatment and supplementation. Annales de Zootechnie. 45 89-103.

 

Wolter R 1994 Alimentation de la vache laitière, Edition France Agricole, Paris, 209 p.



Received 30 May 2003; Review process only began 1 September 2004, due to the paper being mislaid; Accepted 21 September 2004

Go to top