Angela Avery, Agriculture Victoria Rutherglen

The most comprehensive and recent data on the requirements for broodmares and foals is published in the NRC (1989). The authors of this publication acknowledge that for some nutrients, recommended levels of intake are difficult to give due to a lack of published information. NRC (1989) also provides the nutrient content of a number of pasture species, conserved forages, grains and meals. Whether the nutrient concentrations in these species as presented by NRC (1989) apply under Australian conditions needs further consideration.

The number of studies on horse performance on a pasture diet in Australia has been very limited. Gallagher and McMeniman (1988) conducted a trial from June to October in 1985 and Martin (1993) conducted a second trial from July to January in 1986 and 1987 to examine the nutrient intake and reproductive performance of Thoroughbreds and Australian stockhorses grazing pasture in south-east Queensland. The pasture was composed of the following species; blue couch (Cyndon dactylon), Queensland blue grass (Dicanthium sericeum), lotononis (Lotononis bainesil), white clover (Trifolium repens), Rhodes grass (Chloris gayana) and paspalum (Paspalun dilatum). Pasture intake was calculated from estimates of diet digestibility and faeces output which were determined using the internal marker, acid insoluble ash and the external marker chromic oxide, respectively. Samples of diets consumed by these horses were obtained by plucking plants or plant parts that the horses were observed to eat.

The results from the Gallagher and McMeniman (1988) study showed that pregnant mares gained approximately 35 kg/head from July to October. Martin (1993) calculated that nutrient intakes were in excess of NRC (1989) requirements in all but two times of the year; these being in November 1986 when CP was below the level required and in October 1987 when both DE and CP intakes were less than required (table 1).

Gallagher et al. (1994) reported that DE contents of diets selected by horses grazing a mixed oats (Avena sativa), vetch (Vicia atropurpurea) pasture in August and September were 17.2 and 15.4 MJ/kg DM. These results would indicate a dry matter digestibility of 93%, which is very high. Errors in these estimates may have arise due to an inability to select a diet similar to that consumed by the horses and small analytical errors which would lead to large errors in calculated digestibility values.

The energy and protein requirements for broodmares and growing horses as defined by the NRC (1989) are presented in table 2.

The nutritional requirements of broodmares have received limited research effort, with the findings being constrained by being from a different country and limited number of horses. Spring-born fillies normally begin ovulating between 12 to 15 months of age. The commencement of ovulation has been delayed by 2 to 7 months when pony fillies weight remained constant over their first winter (Ellis and Lawrence 1978a). The effects of restricted filly growth on the commencement of ovulation for common breeds in different environments require further investigation.

Mares should have at least a condition score of 5 (Henneke et al. (1983 method)) at the beginning of the breeding season to ensure they commence cycling as soon as possible (Henneke et al., 1984; Kubiak et al., 1987). If mares are thin (CS less than 3) their nutrient intake should be increased 2 to 5 weeks before the beginning of the breeding season so that they are on an upward plane of nutrition (Voss and Pickett, 1974; Kubiak et al. 1987). Morris et al. (1987) have shown that reducing food intake of fat (CS greater than 5.75) mares just prior to the breeding season has an adverse effect on the onset of oestrus.

Pregnant mares

The energy requirement of pregnant mares significantly increases in the final months, 110%, 113% and 120% of maintenance for the 9^th, 10^th, and 11^th months respectively. Nutrient requirements during lactation will be governed by milk yield but can be twice that of the mare’s maintenance requirements. The condition of the mare at foaling can also effect ovulation and conception in the following season, with mares in low condition scores (3.4 or less) having longer post-partum intervals to ovulation, increased cycles per conception and reduced conception rates. The conception rates of mares that were in good condition at foaling (CS 7.5) and lost weight during lactation were similar to those that were in good (CS 7.7) and poor (CS 3.8) condition and maintained or gained weight over lactation (Henneke et al.,1984). Henneke et al. (1984) recommend a condition score of at least 5 at foaling. Several studies have attributed foetal resorptions and early and late abortions to nutrient status of the mare (Merkt and Gunzel, 1979; Mitchell and Allan, 1975; van Niekerk, 1965; Keiper and Houpt, 1984, Merkt and Gunzel, 1979; and Henneke et al. 1984).

Peak lactation in the mare occurs between day 30 and day 70 of lactation, with yields being recorded between 15 and 18 kg/day for 500 kg mares during the first 40 days of lactation (Offedal et al., 1983, and Doreau et al. 1990, 1992). Due to the high energy demand at this time it has been recommended that the mares diet is made up of 30-40% concentration in addition to pasture. Several studies (Martin 1993; and Jordon 1979) have shown that provided pasture availability and nutrient content of the pasture are adequate (DE content of 10 MJ/kg DM and a pasture consumption rate of 12 kg pasture/day for a 500 kg mare producing 18 l/day of milk (NRC, 1989)) lactation can be supported by pasture alone.

Young horses

For most breeds of horses growth is rapid in the first six months and then progressively slows until the horse reaches maturity, which frequently occurs at the age of 5 years. There is increasing pressure in the stud industry to accelerate the growth of young horses to ensure they are well grown and in good condition for sale. There is evidence to suggest that rapid growth early in the horse’s life can have negative effects, especially on the skeleton. Despite the importance of early growth rate in horses there is little information that supports an optimum growth rate for different stages of maturity.

Doreau et al. (1988) determined that foals gained 1 kg of liveweight when fed 9, 13, and 15 kg of milk at week 1, 4 and 8 respectively. In Australia Martin (1993) observed similar growth rates with Australian Stockhorses. It has been shown that mares when fed sub-optimal diets that contained adequate protein were able to supply milk to support adequate foal growth through mobilising body fat stores (Henneke et al., 1984; Pagan et al., 1984). Conversely, mares in poor condition at foaling can maintain adequate milk production providing they have sufficient good quality feed.

Satisfactory growth post-weaning is possible on pasture as long as pasture quality and quality are adequate. Several studies from other countries have investigated post-weaning growth. While this growth is generally adequate, growth rates are often poor over the winter when pasture availability is low (Dawson et al., 1945; Ellis and Lawrence, 1978a,b; and Bigot et al., 1987) in the winter. The information published on post-weaning growth in Australia is extremely limited.

The extensive on-stud measurements taken by Green (1969, 1972) in England and Hintz et al. (1979) in Canada indicate that horses attained approximately 83%, 92%, 95% and 97% of their adult height at the ages of 6, 12, 18 and 22 months old respectively (assuming that the mature size of a stallion is 162 cm and that of a mare is 160 cm). In terms of liveweight these foals reached 46%, 65%, 83% and 90% of their adult weight at 6, 12, 18 and 22 months of age (assuming the mature liveweight of stallions and mares are 545 and 500 kg respectively). The liveweight gain of the foals in Hintz et al. (1979) study was similar to those recommended by NRC (1989) for moderate growth (0.65, 0.50 and 0.35 kg/day for 6, 12 and 18 month colts respectively). Excessive growth in young horses has been associated with developmental orthopaedic disease (DOD) (Glade and Belling, 1984; Thompson et al., 1988; Cyanbaluk et al., 1990; and Savage, 1991).

Protein requirements have been defined by NRC (1989) (table 2) and highlight the importance of the correct lysine to protein ratio. It has been shown by Pulse et al. (1973) that a diet deficient in protein can reduce intake and growth. Published literature indicates that excessive protein appears to have no detrimental effect on horse growth (Yoakam et al., 1978; and Meakim et al., 1981).

Current research on horse performance from pasture in Australia

1.    Subtropical pastures: A comparison of the nutrition of weanlings using set stocking of pasture, rotational grazing and hand-feeding. N. P. McMeniman, The University of Queensland.

In 1995, the Rural Industries Research and Development Corporation (RIRDC) funded a project to compare the growth of hand-fed and grazing horses. The horses used in the study were those weaned in 1996 and 1997 from Australian Stockhorse mares. The growth of these horses was monitored from weaning until they were two years old. For approximately six months in 1997 horses weaned in both 1996 and 1997 were present in the trial.

The horses were divided into three groups of five; one group was hand-fed (50% commercial horse pellets (30 g/head/day), 50% grass hay and a vitamin/mineral mix) to grow at 0.7 kg/day, the second group was rotationally grazed through three paddocks every three weeks (total area of grazing was four hectares) and the third group of animals were continuously grazed in a four hectare paddock. The effective stocking rate for the grazed treatments were one horse to 0.8 hectares except in 1997 where it was one horse to 0.4 hectares.

The horses were grazed on a pasture that was based upon a mixture of kikuyu, green panic, setaria, paspalum, couch and white clover. In late 1995 approximately one quarter of each paddock was prepared and planted with a mixture of Callide Rhodes grass and wyncassia. In the early winter of 1996 and 1997 a mixture of ryegrass, oats and white clover was sod seeded into approximately half the area of each paddock.

The horses were weighed every three weeks, wither heights were measured every three months, the left knee joint of the first two groups of horses were x-rayed when the horses were 12 and 18 months old. Blood samples were obtained periodically for haematological analysis. Pasture yield was measured every three weeks and samples of the pasture on offer were obtained and analysed for nutrient content.

Project results and discussion

Over the experimental period pasture quantity exceeded 1-2 t DM/ha except in the late winter and spring period in 1997. The crude protein of the pasture on offer was consistently at or above 80 g/kg DM. NRC (1989) recommends that the CP content of the diets of horses aged between 6 to 24 months should be in the range of 100-145 g/kg. Due to the grazing behaviour of horses the diet selected would have been higher in protein than the average of the pasture on offer. The researchers suggest that the diet would have therefore contained sufficient protein. Calcium and phosphorus concentrations in the pasture-on-offer ranged from 2 to 3 g/kg DM and 2 to 5 g/kg DM respectively. Unless the horses were selectively grazing for high calcium pasture their calcium intake would have been below requirements and the Ca:P ratio in their diets would have also been low. The only other mineral that was deficient in the pasture was copper.

The liveweight changes of the first group of foals are shown in figure 1. The rotationally grazed horses maintained weight from the 55^th to the 88^th week, while the set stocked horses lost weight from the 50^th to the 70^th week. This period of liveweight maintenance or loss was during the winter in 1997 when pasture availability was low. During this period it was necessary to supplement horses with 2 kg of Pony horse pellets per head per day to minimise weight loss. The final weight of the hand-fed horses were higher than that of the rotationally grazed horses which were higher than the set stocked horses. The skeletal size of the horses reflected by the height at the wither were the same across treatments. The researchers indicate that the primary reason for the difference in the weight between the treatments was body fat (the body condition scores were 4.5 for the hand-fed horses and 3 for the pasture fed horses). The growth curves of the second experimental group were similar to the first experimental group.

Figure 1. Liveweight of the first experimental group (1996-97)

The preliminary conclusion drawn from this experiment was that a subtropical pasture grazing systems, while not supporting the levels of liveweight gain that can be achieved by with full hand- feeding, can support acceptable rates of gain.

In 1996, the Rural Industries Research and Development Corporation (RIRDC) funded a project to compare the growth of hand-fed and grazing horses on temperate patsures. The experiment was conducted on the Rutherglen Research Institute in north-east Victoria. The long-term annual rainfall for the Institute is 600 mm with the rainfall occurring predominantly in winter. The horses used in the study were loaned Thoroughbreds from the foal crops of 1997 and 1998. In 1998, filly foals and in 1999 colt foals were weaned onto the experiment in March and were retained on the plots until December in 1998 and November in 1999. The growth of these horses was monitored from weaning until they were approximately 12-15 months of age.

Each year the experiment consisted of 36 foals, two foals were allocated per treatment to ensure that total liveweight per treatment was as similar as possible. The treatments consisted of improved perennial and unimproved annual pasture at high and low stocking rates compared to foals that were hand-fed on a commercially designed ration to achieve moderate growth rate (0.65-0.70kg/day). The pasture treatments were replicated four times and the hand-fed treatment was replicate twice. In 1997 the improved pasture was limed (2 t/ha) and sown to phalaris and sub clover.

The improved pasture consisted predominantly of phalaris, annual ryegrass and sub clover while the unimproved pasture consisted of annual ryegrass, barley grass, brome and sub clover. The horses were measured every four weeks for height at the wither and hip, liveweight, metacarpal circumference, girth circumference and for intestinal worm eggs. The pasture was measure at the same interval for feed-on-offer, botanical composition, percentage green and nutritive value.

Results presented in this paper are for the first year as the second year was not completed at the time of writing this paper. In addition nutritive value of the pasture samples have yet to be completed. The average weight gain over the experimental period across all pasture treatments was 0.75 kg/head/day compared to the hand-feeding treatment which was 0.69 kg/head/day. Growth rates of 0.78, 0.79, 0.68, 0.73, 0.69 kg/head/day for the improved, high and low stocking rates, unimproved high and low stocking rates and the hand-fed horses were measured respectively (figure 2).

Weight gain of the hand-fed horses was more uneven with this group being the only animals to loose weight but also the group to have the period of highest gain. Weight gain was the lowest in July for all treatments; this is the time of cold wet conditions as well as the period of low pasture growth particularly for the unimproved annual pasture. If it is assumed that the liveweights of mature Thoroughbred mares are 500 kg the foals commenced and finished the experiment with the following percentage of mature weight for improved, unimproved and hand-fed respectively; 37 and 77%, 37 and 72%, and 41 and 76%. These results compared favourably to the rfigures measured from studs in Canada (Hintz et al., 1979), where the percentage of mature weight of 12 and 18 month Thoroughbreds were 65 and 83%.

The gain in height over the experimental period did not vary greatly between treatments with the gains being 18.0, 16.7 and 18.3 cm for the improved pasture, unimproved pasture and hand-fed treatments respectively (figure 3).

Figure 3. Height (cm) of the first experimental group (1998)

 Hintz et al. (1979) also compared the height of Thoroughbred weanlings to their mature height (160 cm for mares) and determined that horses in Canada reached 92 and 95% of their mature height at the ages of 12 and 18 months. In the first year of the experiment the horses on improved, unimproved and hand-fed treatments commenced in March at 79, 80 and 82% of their mature height and finished in December at 90, 90 and 93% of their mature height.

The preliminary conclusion of this work is that temperate pasture can support acceptable horse growth over the autumn to early summer in temperate south-eastern Australia. It is however important to recognise several factors; (1) there was a large amount of standing pasture on the paddocks at the start of the experiment; (2) that the time of greatest pasture shortage (quantity and quality) in this environment is from January to March; and (3) this experiment was conducted in a harsher environment than in those regions where horse studs tend to be situated.

Bigot, G., Trilland-Geyl, C., Jussiaux, M. and Martin-Rosset, W. (1987). Elevage du cheval =de selle du severrage an debourrage: alimentation hivornale, crossance et developpment. Bull technique Centre de Recherches Zootechiques et Veterinaires de Theix. No. 69, pp 45-53.

Cymbaluk, N.F. and Christison, G.I and Leach, D.H. (1990). Longitudinal growth analysis of horses following limited and ad libitum feeding. Equine Vet. J., 22 : 198-204.

Dawson, W.M., Phillips, R.S. and Speelman, S.R. (1945). Growth of horses under western range conditions. J. Anim. Sci., 4: 47-54.

Doreau, M., Boulet, S., Barlet, J.P. and Patureau-Mirand, P. (1990). Yield and composiion of milk from lactating mares: effect of laceration stage and individual differences. J. Dairy Res., 57: 449-454.

Ellis, R. N. W., and Lawerance, T. L. J. (1978b). Energy under nutrition in the weanling filly foal. II. Effects on heart rate and subsequent voluntary food intake. Br. Vet. J., 134: 333-341.

Gallagher, J. R. and McMeniman, N.P. (1999). The nutritional status of pregnant and non-pregnant mares grazing South East Queensland pastures. Equine Vet. J., 20: 414-416.

Gallagher, J. R., McMeniman, N.P., Reeves, T.G. and Washington, D. (1994). The digestible energy and nitrogen intake of horses grazing oat/vetch pastures. Proc. Aust. Soc. Anim. Prod., 20: 410.

Green, D.A. (1969). A study of growth rate in Thoroughbred foals. Br. Vet. J., 125: 539-546.

Green, D.A. (1976) Growth rate in Thoroughbred yearlings and two year colts. Equine Vet. J., 8: 133-134.

Henneke, D.R., Potter, G.D., Kreider, J.L. and Yeates, B.F. (1983). Relationship between condition score, physical measurements and body fat percentage in mares. Equine Vet. J., 15: 371-372.

Henneke, D.R., Potter, G.D., Kreider, J.L. and Yeates, B.F. (1984).Body condition during pregnancy and lactation and reproductive efficiency of mares. Therio., 21: 897-909.

Hintz, H.F, Hintz, R.L. and Van Vleck, L.D. (1979). Growth rate of Thoroughbreds, effect of age of dam, year and month of birth and sex of foal. J. Anim. Sci., 48: 480-487

Jordon, R.M. (1979). A note on the energy requirements for lactation of pony mares. Proc, 6^th Equine Nutr,. Physiol. Symp. Texas A&M University, Texas. pp. 27-32.

Keiper, R. and Houpt, K. (1984). Reproduction in feral horses: an eight year study. Am. J. Vet. Res., 45: 991-995.

Keiper, R., Crawford, B.H., Squires, E.L., Wrigley, R.H. and Ward, G.M. (1987). The influence of energy intake and percentage of body fat on the reproductive performance of non pregnant mares. Therio., 28: 587-598.

Martin, R.G. (1993). Effects of nitrogen and energy supplementation on grazing broodmares and aspects of nitrogen metabolism in horses. PhD Thesis, The University of Queensland.

Meakim, D.W., Hintz, H.F., Schryver, H.F. and Lowe, J.W. (1981). The effect of dietary protein on calcium metabolism and growth of the weanling foal. Proc. Cornell Nutr. Conf., pp 95-102.

Merkt, H. and Gunzel, A.R. (1979). A survey of early pregnancy losses in West Gggggerman Thoroughbred mares. Equine Vet. J., 11: 256-258.

Mitchell, D. and Allen, W.R. (1975). Observations on reproductive performance in the yearling mare. J. Reprod. Fertil. Suppl., 23: 531-536.

Morris, R.P., Rich, G.A., Ralston, S.L, squires, E.L. and Pickett, B.W. (1987). Follicular activity in transitional mares as affected by body condition and dietary energy. Proc. 10^th Equine Nutr. Physiol. Symp. Colarado State University, Colorado, pp. 93-99.

Oftedal, O.T., Hintz, H.F. and Schryver, H.F. (1983). Lactation in the horse: milk composition and intake by foals. J. nutr., 113: 2093-2106.

Pagan, J.D and Hintz, H.F. (1986). Composition of milk from pony mares fed various levels of digestible energy. Cornell Vet., 76: 139-148.

Thompson, K.N., Jackson, S.G. and Baker, J.P. (1988). The influence of high planes of nutrition on skeletal growth and development of weanling horses. J. Anim. Sci., 66: 2459-2467.

Van Niekerk, C.H. (1965). II. Early embryonic resorption in mares (a preliminary report). J. S. Afr. Vet. Med. Ass., 36: 61-69.

Voss, J.L. and Picket, B.W. (1974). Effect of a nutritional supplement on pregnancy rate in non-lactating mares. J. Am. Vet. Med. Ass., 165: 702-703.

Yoakam, S.C., Kirkham, W.W. and Beeson, W.M. (1978). Effect of protein level on growth in young ponies. J. Anim. Sci., 46: 983-991.