Livestock Research for Rural Development 6 (2) 1994

Citation of this paper

Effect of work on dry matter intake, milk production and reproduction in multipurpose cows fed low quality roughage

Ercole Zerbini* and Takele Gemeda**

* International Livestock Centre for Africa (ILCA), Addis Ababa, Ethiopia. ** Institute of Agricultural Research (IAR), Addis Ababa, Ethiopia.

(This paper was presented at the "International Conference on Increasing Livestock Production through Utilization of Local Resource". CECAT, Beijing, 18.22 October 1993)


The use of multipurpose animals for milk meat and draught, has shown potential for increasing farm productivity and income. Females used for draught require better quality feed and management than oxen or dry cows. Despite the requirement for higher quality feed, there is potential for saving feed resources overall because fewer animals are needed to supply total farm work and milk needs. Research has shown that conception and milk production of cattle and buffaloes are affected if diets based on crop residues or low quality hays are not improved or supplemented, especially when working periods coincide with lactation. Our research showed that hay (~7.0 MJ ME/kg DM) intake was 10% greater in working compared to non.working cows. Cows responded to underfeeding by a progressive adaptation until a new equilibrium between dietary supply and requirements was reached. However, in draught cows fed low quality roughage, the "work.induced" increased dry matter intake was not sufficient to cover all energy requirements for work and/or milk. While inadequate feeding may not affect work output, high rates of body weight loss lead to decreased milk production and reproductive ability. Total milk production in a two.year period was similar for working and non.working cows (1985 vs. 2225 kg). Work had a minimal affect on milk production when nutrition was adequate for lactation and work. Roughage diets with less than 8 MJ ME/kg DM could not support potential reproductive ability of draught cows. Supplementation improved reproductive performance of working cows and by 365 days post.partum, conception rate of non working.supplemented and of working.supplemented cows was similar. However, in working cows, conception was delayed by about 120 days. Body condition at calving significantly affected post.partum reproductive ability of cows. Many more site.specific investigations need to be carried out to find the most economic combinations of feeds throughout the year and to relate food quality and intake to animal performance, especially in early lactation when the high energy demand for lactation could be associated with work energy needs.

KEY WORDS: Work, draught, multipurpose, cattle, milk, reproduction, roughages


Where animal traction is currently used, the major sources of fodder are crop residues, conserved forage and browse. The intensification of agriculture leaves fewer and fewer opportunities for animals to graze, and feed resources that are alternative to "conventiona"' grazing are becoming increasingly important (Mathers and Otchere 1993).

Energy expenditure and feed intake

In animals used for draught, maintenance and work energy are quantitatively the most important. Apart from extra net energy for work, draught animals also require more energy for maintenance than non.working animals. Work by Lawrence et al (1989a,b) shows an increase of about 10% in net energy for maintenance of working oxen fed poor quality diets.

Even though it is now possible to define many of the energy expenditures of draught animals reasonably well, there are still difficulties in estimating the metabolizable energy content of many tropical feeds and in predicting the voluntary intake of low quality roughages necessary to provide the metabolizable energy to meet these requirements (Lawrence 1987).

The main factor which determines intake is food quality, but other factors, such as ambient temperature and physiological state of the animal are also important. Even though published data have shown a strong relationship between DM intake and production responses, it has been difficult to establish cause and effect. Weston (1985) suggested that increased energy demand by an animal would result in increased feed intake. While this is true in lactating cows, results of studies on the effect of work on feed intake are inconsistent. Winugroho (1993) indicated that one of the causes of this inconsistency could be attributed to the different efficiency with which available energy is used which depends on the supply of nutrients (feed, body reserves) and on the intensity of work. In working animals, increased heat load and decreased gut motility and rate of passage may reduce appetite and intake (Young 1985). Thus, a major increase in feed intake in working animals exposed to heat stress from work and from the environment will be difficult to achieve. This would be particularly true in Buffalo, which have a much poorer heat dissipating capacity than cattle.

In studies cited by Mathers and Otchere (1993), there was an increase in food intake during work in six of the cases; in four studies, there was a decrease. In most cases, however, the difference was not statistically significant. Henning (1987) reports that the difference between energy demand and energy supply per se is not the major factor in increasing feed intake. In the study reported, feed intake did not increase as a result of increased energy expenditure through walking exercise. Similarly, Bamualim et al (1987) showed no significant differences between working and non.working animals in their feed intake and digestibility, in contrast to previous findings by Ffoulkes and Baumalim (1986) who reported increases in both feed intake and digestibility in buffalo after 3 weeks of work. Matthewman et al (1989) indicated that exercise did not affect dry matter intake of cows.

Based on observations of the nutritional status of working animals, it is suggested that in areas with available roughages containing more than 9 MJ of ME/kg DM, feeding of draught animals should be adequate (ARC 1980; Lawrence 1987). However, many unsupplemented and unmodified tropical forages do not contain that concentration of energy and would be incapable of supporting intakes up to twice maintenance. Feeds which are normally fed to working cattle and buffalo are fibrous roughages and their ME content varies from 6 to 8 MJ/kg DM. Therefore, liveweight losses in animals eating these diets during the work season would be expected. Greater energy deficit would occur if the working animals would be lactating and/or pregnant (Teleni 1993).

Even under conditions where adequate feed supplementation is not available to maintain body weight, animals can still satisfactorily perform work by drawing on body reserves. The length of time such a situation can effectively exist will depend on the previous condition of the draught animal, the amount of work being done and its duration (Goe 1987).

Even though draught animals should be in good condition especially at the beginning of the working periods, work conducted by Khibe and Bartolomew (1993) in Mali shows that feeding for weight maintenance during the dry season would represent an inefficient use of feed resources compared with short term feeding during the working period. Although animals had lost an average of 50 kg during the dry season, there was no evidence of any effect of liveweight or body condition on work output in observations made during the cropping season. Therefore, the economic trade.offs between loss of body condition, work output and costs of feeding supplements during and before the working season should be examined in detail. This issue becomes more complex with working cows which need to maintain body weight after parturition in order to conceive (Zerbini et al 1993).

Utilization of crop residues and browse

A knowledge of the energy requirements of draught animals enables rations to be formulated from any feeds which are available provided their metabolizable energy contents are known. Research tends to focus on how best to make use of the limited feed available and on ways of supplementing or treating feeds to improve their quality (Lawrence and Zerbini 1993). Experiments by Chiadet (1989), quoted by Lawrence and Zerbini (1993), used 4 pairs of buffaloes to test 4 different diets. The untreated diet consisted of rice straw and fresh grass, and in the others, the rice straw was replaced by urea.treated rice straw and/or a concentrate ration. All 3 improved diets resulted in a 7% increase in ploughing capacity, in increased weight gain during rest periods and a diminution in weight loss when the animals worked.

Combinations of urea and concentrate (cotton seed cake) were also used by Khibe and Bartholomew (1993) to improve a basal diet of "bush hay" fed to working oxen. Animals receiving untreated hay or hay + urea gained on average 20 kg during the trial and those receiving hay + urea + concentrate gained 40 kg. The work required of the animals was relatively low and no difference in work performance was noted between the groups. However, the improvement associated with the cotton seed cake may have been due as much to the supplementary metabolizable energy provided as to the supplementary protein. In another study, working oxen receiving only rice straw lost 10 kg liveweight on average during the trial whereas those receiving a supplement of 4 kg/day of fodder tree Ficus Auriculata (21.6% N x 6.25) made minimal gains (Pearson 1991).

The diversity of products that have been used as feeds or supplements for draught animals indicate the need for site.specific research to ensure their efficient utilization.

Nutrition of multipurpose draught animals

Animals are seldom kept exclusively for draught; they are often trying to grow or lactate at the same time. Much recent research has focused on the partition of nutrients in the animal between these various functions and the best ways of formulating diets to meet nutrient needs.

The opportunity costs for farms that maintain oxen primarily for land cultivation are considerable and are likely to become greater as human and livestock pressure on land increases. Draught oxen also compete with cows and associated stocks for normally limited supplies of foodstuffs. The use of crossbred dairy cows for the dual purpose of milk production and draught is one possible means of alleviating this situation, especially in areas where dairy production schemes are being implemented in mixed cereal.livestock farming systems (Gryssels and Anderson 1985). In many tropical farming systems, the preferred draught animals are oxen, but cows are used where available land and feed for cattle are scarce. For instance in Bangladesh, Indonesia, Sri Lanka, and Senegal between 25 and 50% of draught animals are cows. However, other situations exist where land and population pressure associated with inefficient use of draught oxen is critical to crop production (Matthewman et al 1990). Draught cows are used primarily where feed resources for draught animals are scarce. However, feed shortage is not the only factor governing the choice between using oxen or cows. Other factors include: shortage of males, human population pressure, sub.division of holdings, and shortage of capital (Jabbar 1983).

Draught cows have higher nutrient requirements than oxen especially if they have to perform draught work during the early stages of lactation when nutrient supply has to cover the needs for work, lactation, and reproductive activity. Therefore it is especially important that cows to be used for draught be in reasonable body condition all year round. This will allow them to sustain all functions required without undue loss of body weight and a limited effect on milk yield and reproductive performance. Supplementary feeding may be necessary unless good quality forage is available all year round.

Uncertainty exists at the present time about the exact effect of work on milk production and, while some authors consider that well fed cows suffer no adverse effects from working, it would appear that milk yields and fertility are affected in poorly fed animals.

Matthewman (1990) indicated that dairy cows using approximately 12 MJ ME per day for work showed a reduction of milk production between 7 and 14% depending on diet. However, milk yield declined on walking days and recovered to almost non.walking levels on the intervening non.walking days. The daily yield of milk fat, protein and lactose were not affected by exercise. Body weight change was similar. The author concluded that the amount of exercise done by the cows would not have any long term effects on lactation performance.

Energy metabolism and partition in dairy cows used for draught There is a need to quantify the energy partition to different functions by the working.lactating.conceiving cow. How body reserves contribute to energy expenditure of working cows is not known satisfactorily and there appears to be no direct information on the extent to which work might influence the composition of gain or loss. The need for the animal to regain body weight during the non working season is a further feature of draught animal management.

There is a general consensus that free fatty acids and glucose are the nutrients preferentially used by muscles of working animals (Teleni and Hogan 1989). Glucose and NEFA are key players in the energy metabolism of lactating cows. Glucose is not only a major substrate for lactose synthesis, but also an essential energy substrate for reproductive activity (Gaines 1989; Butler and Smith 1989; Harrison 1982). Fatty acids are precursors of milk fat, but also seem to influence neuroendocrine activities regulating the synthesis and release of reproductive hormones (Shillo 1992). Thus, a strong competition for these metabolites could occur when cows are lactating and working during the same period. Intermediate metabolites of carbohydrate and lipid metabolism such as lactate and ketone bodies would also play an important role in the overall energy partition strategy in the lactating working cow. The specific effect of work on energy metabolites must be an important factor in evaluating the overall efficiency of cows used for draught.

Work carried out at the International Livestock Centre for Africa (ILCA 1990) has shown that non.supplemented cows seemed to have partitioned the available energy towards maintenance and work, penalizing milk production and reproduction. Among supplemented cows, those working exhibited only a greater postpartum interval from calving to conception. It is not known whether it was a direct effect of work per se or a deficit of energy yielding substrates, in particular during working periods.

Data from the ILCA/IAR cow traction research project (Zerbini et al, unpublished) show that plasma glucose was greater in supplemented than in non.supplemented cows in both working and non. working cows (8 to 16%). Working cows had lower plasma glucose after working hours than non.working cows (.16%). NEFA was higher in working cows at the end of working hours and after the rest hour between working hours. This effect was greater in non.supplemented than in supplemented cows. The rise of plasma NEFA of about 150% in working cows during working hours is in agreement with the results of Hambitzer and Bent (1988) which showed that free fatty acids increased 7.fold in exercised Arab horses. Similarly, Teleni et al (1989) reported an increase in glucose entry rate of approximately 84% and an increase in FFA entry rate of approximately 150% in working cattle. In the pregnant and/or lactating working cows or buffaloes the increased entry rate of glucose could be accounted for by the gravid uterus or the mammary gland uptake. In our study, plasma glucose and NEFA showed a negative linear relationship in non.working/supplemented, working/non.supplemented and working/ supplemented cows. The decrease of plasma glucose and the increase in NEFA during work indicates a declining utilization of glucose and an increasing use of NEFA by muscle during consecutive working hours. Working/supplemented cows might have used dietary fat and fat and carbohydrate stores in muscle to a larger degree than working/non.supplemented cows.

The results of this study and those reported by Zerbini et al (1993) for the same cows, suggest that an average decrease of plasma glucose from 55.5 to 50.0 mg/dl over all sampling times could be related to a decrease of the probability of conception by a factor of 4 in non.supplemented compared to supplemented cows. On the other hand, an average reduction of plasma glucose from 54.9 to 46.0 mg/dl, and an increase of NEFA of 146% in working cows during working hours could be related to a decrease of the probability of conception by a factor of 2 for both diets. Similarly, the lower plasma glucose of non.supplemented cows compared to supplemented cows relates well with the effect of diet supplementation on milk yield reported by Zerbini et al (1992) with the same cows. Milk production over one.year period was reduced by 53% in non. supplemented cows. During the same period the effect of work was not significant.

Energy requirements of the working cow

When cows are used for work, the additional energy requirement for work must be added to maintenance and milk production (Table 1).

Table 1: Calculated daily energy requirements of crossbred cows*
Work output = 2.7 MJ/day; body weight = 420kg,  
Milk yield = 4.8 kg/d.  
ME MILK (4.8 kg/d) 25.4 MJ

(*) Feeding Standards for Australian Livestock 1990)

At the end of the first working period after calving (90 days), hay dry matter intake of non supplemented.working cows increased by 1.6 kg/day compared to non supplemented.non working cows, equivalent to 12 MJ ME associated with a body weight loss of 0.39 kg/day which equates approximately to 14.4 MJ ME. The total output was there fore 26.4 MJ ME per day. Since the total energy requirements of working cows are 98.1 MJ/d they would suffer a deficit of 12.1 MJ/day which would affect milk yield and body weight considerably. Non working cows showed a deficit of only 5.3 MJ per day.

Production and reproduction performance of crossbred dairy cows used for draught

The following results show the effect of work and diet supplementation on diet apparent digestibility, dry matter intake, milk production and body weight change over a period of two years. This period included 200 days of work.

Feed intake, feed utilization and work output

Table 2 shows preliminary results of in vivo digestibility of organic matter and calculated ME content of hay and supplemented diets comparing non working cows and working cows during a working period. Supplementation increased organic matter digestibility by 4.4 % units in non working cows and by 3.3 % units in working cows. The digestible organic matter in feed dry matter was greater for supplemented cows but differences were not as significant as for organic matter digestibility. Calculated metabolizable energy reflect the greater value of the supplemented diet, but also the greater feed utilization of working compared to non working cows. Similarly, Ffoulkes and Baumalim (1989) reported that work increased the availability of feed energy in buffalo.


Table 2: Organic matter in vitro digestibility (ODM), digestible organic matter in feed dry matter (DOMD and calculated ME of diets fed to F1 crossbred dairy cows used for draught. 2. Non.Working cows vs. Working cows during working periods.
  (%) (%) (MJ/kg DM)
NW.NS** 54.7 49.1 7.37
NW.S 59.1 52.7 7.91
W.NS 57.9 53.0 7.95
W.S 61.2 55.4 8.31
ąSE 2.0 1.7  
Work NS < 0.06  
Supplement <.05 < 0.08  

* ME = 0.15 x DOMD (MAFF 1984)
** NW.NS = Non Working . Non Supplemented
NW.S = Non Working . Supplemented
W.NS = Working . Non Supplemented
W.S = Working . Supplemented

Tables 3 and 4 show that over one and two year periods, total hay dry matter intake was greater for working than non working cows and for supplemented than non.supplemented cows. Supplemented cows consumed more dry matter than non.supplemented cows. Total dry matter intake followed a similar pattern. Results indicate that working cows increased dry matter intake compared to non working cows to sustain milk production and/or support energy expenditure for work. Of particular practical importance is the increased hay dry matter intake of non.supplemented working cows. Ruminants respond to underfeeding by a progressive adaptation until a new equilibrium between dietary supply and requirements is reached. Cows in our study increased dry matter intake to partly support greater energy demands for work. However, as indicated by Ortiguez (1991), there might be a drop in maintenance energy requirements originating from a decrease in basal metabolism that depends on the level and duration of undernutrition. If this could be the case for non working cows, it probably would not hold for working cows which increased dry matter intake, thus increasing digestive tract metabolism. A desirable trait of the lactating working cows would be a large food intake capacity. Larger animals could be of considerable advantage in situations where high fibre roughages are utilized. Larger animals are more efficient chewers and spend less time chewing per kg of ingested cell wall constituents. Statistically, body size accounts for 52% of variation in intake (Bae et al 1983).


Table 3: Work output, dry matter intake, milk yield and body weight change of F1 crossbred dairy cows used for draught over a period of 365 days.
  Intake Intake Milk Bodyweight
  Hay DM yield change
  (kg) (kg) (kg) (kg)
NW.NS 2559 2559 860 .59.8
NW.S 2950 3755 1764 9.4
W.NS 3020 3020 801 .61.0
W.S 3207 4139 1756 25.2
ąSE   141 150 9.1
Supp P<.05 P.001 P<.001 P<.001

* P < 0.05 ** P < 0.01 *** P < 0.001


Table 4: Work output, dry matter intake, milk yield and body weight change of F1 crossbred dairy cows used for draught over a period of 730 days.
  Intake Intake Milk Bodyweight
  Hay DM yield change
  (kg) (kg) (kg) (kg)
NW.NS 5164 5164 1263 .70.0
NW.S 5676 7093 2186 .12.4
W.NS 6055 6055 927 .85.6
W.S 6358 7957 3044 .23.4
ąSE   241 216 9.1
Work P<.001 P<.001 NS NS
Supp P.05 P<.001 P<.001 P<.001


Work output of supplemented and non.supplemented cows was similar in the first and second year periods (260 vs. 270 and 229 vs. 248 MJ, respectively).


During the first year, the effect of work on average daily milk yield was not significant even though milk yield was reduced by work during the first working period (90 days). However, over both years, total milk production was similar for working and non. working cows. Supplemented cows produced significantly more milk than non.supplemented cows (Table 3 and 4).

Non.supplemented cows lost body weight throughout the two.year period. Supplemented cows, except for working cows during the first and last working periods (90 and 540 days), maintained initial body weight. Total body weight losses were similar for working and non.working cows (Table 3 and 4). Total body weight change of non.supplemented cows was significantly greater than that of supplemented cows. Similar body weight losses in working and non.working cows and minimal body weight loss of supplemented working cows compared to non.supplemented working cows indicated that the "work.induced" increased hay dry matter intake in non. supplemented working cows was insufficient to cover all energy requirements. Thus, non supplemented cows continued to loose body weight and stopped milk production and reproduction functions as a result of work. On the other hand, supplemented working cows were able to work and produce milk without adverse effect on body weight.

Table 5: Numbers of cows which showed oestrus or conceived by 365 days and median intervals from calving to conception
      Oestrus   Conception  
      Median   Median  
      interval   interval  
Treatment n n (days) n (days)  
S.NW 10 10 76 10 76  
S.W 10 10 198 9 227  
NS.NW 10 10 153 6 251  
NS.W 10 2 NA 2 NA  
All 40 28 211 27 235  

Source: Zerbini et al 1993


Body weight losses greater than 15% have been reported to impair ovarian activity in female buffaloes (Teleni et al 1989). Work carried out in early lactation could delay return to oestrus. Reduced blood sugar levels resulting from work could affect implantation if work is carried out at this time. Reducing body weight loss from 0.7 to 0.3 kg/day with supplementary feeding has been shown to improve reproductive performance by about 65% (ILCA 1990). Table 5 shows the effect of work and diet supplementation on conception rate of cows used for draught. In supplemented cows, work significantly delayed days to conception (Zerbini et al 1993). However, by 365 days post.partum, conception rate was similar for supplemented.working cows and supplemented.non working cows. Body condition at calving significantly affected post.partum reproductive ability of non working and working cows. The incidence of ovulation without estrus was higher in working than in non. working cows and in non supplemented than in supplemented cows. Even though in working cows diet supplementation offset the negative effect of work on the onset of estrus and conception, a relatively higher number of cows in the supplemented working group had ovulations without estrus before a normal estrus cycle was established (Zerbini et al 1993).

Natural grass hay alone could not support potential reproductive ability of crossbred cows. Work output of supplemented cows might be associated with longer calving intervals. The economic trade.offs between longer calving intervals and work output should be examined in detail.

It is unlikely that the cessation of cyclic activity in working cows was a result of direct competition for nutrients between the ovary and other tissues. It is possible that the depletion of body reserve nutrients to certain critical levels had signalled metabolic controls to switch off non vital processes such as ovarian function. A clear definition of body weight and condition at the start of the work season and rate of weight loss which are compatible with normal ovarian activity is desirable, as well as the effect of interaction between work and body reserve nutrients on cyclic activities in cattle and buffaloes. Work did not influence conception rate in supplemented cows, but had a substantial influence in non.supplemented cows. The availability of nutrients, particularly, energy and protein, is often a limiting factor for production in cows fed crop by products. Nutritional constraints combined with draught work activities could cause a loss in body condition and lead to postpartum anestrus (Jainudeen, 1985). Therefore, specific intervention options on the amount of work and/or feeding regimes sufficient to induce behavioral estrus during the early postpartum period need to be established.

The significant effect of body condition at calving on onset of oestrus and conception, indicates the importance of the partition of body energy stores and feed energy for different physiological functions when cows are simultaneously working and lactating. Future research should include defining priorities of nutrient partition in lactating and/or pregnant working females. Detailed physiological studies on post.partum reproductive functions in relation to hormonal changes, nutrition, work and their interaction, is warranted.

Economic appraisal of effect of work and diet Partial budgeting analysis was used in the economic assessment of effects of work and diet supplementation on dairy cows on.station. Table 6 shows that regardless of work treatment, gross margin from supplemented cows was much higher than from non supplemented cows which in fact resulted in negative gross margin. The higher feed costs, and lower value for milk and calves of working cows were largely offset by work output value, resulting in higher gross margin for working cows.

This comparison shows the economic viability and the sustainability of the supplemented.working treatment and demonstrates that adequate feeding is a major factor determining the profitability of the introduction of crossbred cows into the farming system, independently of their use for draught work or not (Zerbini and Getachew Asamenew 1991). In addition, non adequate feeding makes the introduction of crossbred cows not a sustainable option because cows not supplemented result in lower calf output and eventually in a complete stop of reproductive functions (Zerbini et al 1993) with loss of production and cow value.


Multipurpose cows could play an important role as dual purpose animals in many ecoregions where smallholder farming systems are established. In particular, their preference over oxen could contribute to a better utilization of feed resources that are already scarce. In addition, the use of cows for draught would allow for males to be fattened and sold younger and to a greater security of replacements. More productive animals on farm could allow for a reduction of stocking rate and overgrazing, thus contributing to the establishment of a more stable soil system. However, a reduction of herd size by using crossbred cows for traction will depend primarily on the available supply of crossbred animals and farmers access to suitable feeds and veterinary services. For feeding practices, two possible solutions may be recommended where crossbred cows are used for draught:

(i) Production and feeding of improved forages (grasses+legumes, legumes, or MPT) to increase digestibility and energy intake of cows to levels which would allow them to support both milk production, reproduction, and work with acceptable physiological body weight loss.

(ii) Feeding of improved quality crop residues and production of well managed natural pasture hays associated with concentrate feeding during early lactation, especially if cows are due to work during that period.

Application of new techniques and research findings for better utilization of crop by.products and conservation of the natural forage during particular periods of the year needs particular attention.


ARC 1980 The nutrient requirements of ruminant livestock. Commonwealth Agricultural Bureaux, Farnham Royal, UK 351p

Bae Dong Ho, Welch J G and Gilman B E 1983 Mastication and rumination in relation to body size of cattle. Journal Dairy Science 66:2137.2141

Bamualim A, Ffoulkes D and Fletcher I C 1987 Preliminary observations on the effect of work on intake, digestibility, growth and ovarian activity of swamp buffalo cows. Draught Animal Power Project, DAP Project Bulletin, No. 3, pp6.10. James Cook University, Townsville (Australia)

Butler W R and R D Smith 1989 Interrelationships between energy balance and postpartum reproductive function in dairy cattle. Journal of Dairy Science 72:767.783

Chiadet P 1989 The effects of urea.treated rice straw and/or feed supplement on voluntary feed intake, digestibility and working capacity in draft buffaloes. MSc thesis, Faculty of Graduate Studies, Khon Kaen University, Thailand

Feeding Standards for Australian Livestock. Ruminants. CSIRO, Australia, 1990

Ffoulkes D and Bamualin A 1989 Improving the nutrition level of draught animals using available feeds. In Draught Animals in rural development. (Editors: Hoffman D, Nari J and Petheram R J). ACIAR Proceedings No. 27, ACIAR, Canberra, pp146.154

Gaines J 1989 The relationship between nutrition and fertility in dairy herds. Veterinary Medicine October 1989, pp997.1002

Goe M R 1987 Animal traction on smallholder farms in the Ethiopian Highlands. PhD Dissertation. Cornell University, USA

Gryssels G and F M Anderson 1985 Use of crossbred dairy cows as draught animals: experience from the Ethiopian Highlands. In: Research Methodology for Livestock on Farm Trials. Proceedingsof a workshop held March 25.28 at Aleppo, Syria. International Development Research Centre, Ottawa. p237.258

Hambitzer R and Bent E 1988 Effect of long distance exercise on free fatty acids in blood plasma of arab horses. Journal of Veterinary Medicine 35:622.625

Harrison R O, Young J W, Freeman A E and Ford S P 1989 Effects of lactational level on reactivation of ovarian function, and interval from parturition to first visual estrous and conception in high.producing Holstein cows. Animal Production 49:23.28

Henning P H 1987 The effect of increased energy demand through walking exercise on intake and ruminal characteristics of sheep fed a roughage diet. Journal of Agricultural Science, Cambridge, 109:53.59

ILCA 1990 Annual Report and Research Programme Highlights. International Livestock Centre for Africa. Addis Ababa, Ethiopia

Jabbar M A 1983 Effects of draught use of cows on fertility, milk production and consumption. In: Maximum Livestock production from minimum land. (Editors: Davis C H; Preston T R; Haque M; Saadullah M). James Cook University: Townsville (Australia) p71.83

Jainudeen M R 1985 Reproduction in draught animals: Does work affect female fertility. In: Draught Animal Power for Production (Editor: J W Copland). Proceedings of an International Workshop held at James Cook University, Townsville, Qld, Australia, July 10.16, pp130.133

Khibe P and Bartholomew P 1993 Dry season feed supplementation for work oxen in the semi.arid zone of sub.sahelian Africa. In Research for Development of Animal Traction (Editors: Lawrence and Dijkman), The proceedings of the 4th Workshop of the West African Animal Traction Network, held 4.9 July 1990 in Kano, Nigeria. International Livestock Centre for Africa, Addis Ababa, Ethiopia

Lawrence P and Zerbini E 1993 Recent trends in research on draught animal nutrition. Paper presented at the first workshop on Human and Draught Animal Power in crop Production: Experiences, Present Status and Research Priorities. January 19.22, 1993. Harare, Zimbabwe

Lawrence P R 1987 Nutrient requirements of working ruminants. In: An Introduction to Working Animals (Editor: J Lindsay Falvey). p61.80

Lawrence P R, Buck S F and Campbell I 1989a The metabolic rate of oxen after work. Proceedings of the Nutrition Society 48:p153A

Lawrence P R, Sosa R and Campbell I 1989b The underlying "resting" energy consumption of oxen during work. Proceedings Nutrition Society, 48:154A

MAFF 1984 Energy allowances and feeding systems for ruminants. Reference Book 433. Ministry of Agriculture, Fisheries and Food. HMSO. London

Mathers J C and Otchere E O 1993 Research on nutrition of working animals: needs, experiences and methods. In Research for Development of Animal Traction (Editors: Lawrence and Dijkman), The proceedings of the 4th Workshop of the West African Animal Traction Network, held 4.9 July 1990 in Kano, Nigeria. International Livestock Centre for Africa, Addis Ababa, Ethiopia (in press)

Matthewman R W 1989 Effect of sustained exercised on milk yield, milk composition and blood metabolite concentrations in Hereford x Friesian cattle. PhD Dissertation. Centre for Tropical Veterinary Medicine, Roslin, Midlothian, EH25 9RG

Matthewman R W, Dijkman J T and Zerbini E 1990 The management and husbandry of male and female draught animals: Research achievements and needs. Paper presented at the fourth Workshop of the West African Animal Traction Network entitled "Research for Development on Animal Traction in West Africa" held on 9.13th July 1990, in Kano, Nigeria

Ortigues I 1991 Adaptation du metabolisme energetique des ruminants a la sous.alimentation. Quantification au niveau de l' animal entier et de tissus corporels. Reproduction Nutrition Development 31:593.616

Pearson R A 1990 A note on liveweight and intake and digestibility of food by draught cattle after supplementation of rice straw with the fodder tree Ficus auriculata. Animal Production 51:635.638

Shillo K K 1992 Effects of dietary energy on control of luteinizing hormone secretion in cattle and sheep. Journal of Animal Science 70:1271.1282

Teleni E 1993 Energy expenditure and nutrient requirement of working animals In: Draught Animal Power in the Asian. Australasian Region. (Editor: Pryor W J) A workshop in conjunction with 6th. Asian.Australasian Association of Animal Production Societies, Congress, 23.28 November, Bangkok, Thailand. ACIAR Proceedings No. 46. pp93

Teleni E and J P Hogan 1989 Nutrition of draught animals. In: Draught Animals in Rural Development. ACIAR Proceedings No 27. 118.133

Teleni E, Boniface A N, Sutherland S and Entwistle K W 1989 The effect of depletion of body reserve nutrients on reproduction in Bos indicus cattle. Draught Animal Power Project Bulletin, 8: 10. August, 1989. James Cook University. Townsville, Australia (E)

Winugroho M 1993 Feeding draught animals in Indonesia. In: Draught Animal Power in the Asian.Australasian Region. (Editor: Pryor W J). A workshop in conjunction with 6th. Asian.Australasian Association of Animal Production Societies, Congress, 23.28 November, Bangkok, Thailand. ACIAR Proceedings No. 46. pp93

Zerbini E and Getachew Asamenew 1991 Potential of cow traction in the Ethiopian highlands farming system. Paper presented at the 4th. National Livestock Improvement Conference, 13.16 November 1991, held in Addis Ababa, Ethiopia Zerbini E, Takele Gemeda, Alemu G/Wold, and Abiye Astatke 1992 Effect of draught work on lactation of F1 crossbred dairy cows. (Preliminary results). Paper presented at the First Workshop of the Animal Traction Network for Eastern and Southern Africa (ATNESA) held 18.23 January 1992, Lusaka, Zambia

Zerbini E, Takele Gemeda, R Franceschini, J Sherington and Alemu G/Wold 1993 eproductive performance of F1 crossbred dairy cows. Effect of work and diet supplementation. Animal Production, 57(3)

Zerbini E, Takele Gemeda, Tegene Azage, Wold Alemu Gebre and Franceschini R 1993 Effects of work and diet on progesterone secretion, short luteal phases and ovulations without estrus in postpartum F1 crossbred dairy cows. Theriogenology, 43(3):571.584