Veld hay as feed for Matebele goats during the last trimester of pregnancy

L M Sibanda ¹, L R Ndlovu ¹ and M J Bryant ²

¹ Department of Animal Science
University of Zimbabwe
PO Box MP 167, Mount Pleasant, Harare, Zimbabwe

² Department of Agriculture
University of Reading
Earley Gate, Reading RG6 2AT, UK

ABSTRACT

Under natural husbandry conditions, kid birth weights and growth rates are lowest in the dry season when feed is scarce and of very low quality. The experiment investigated the potential of veld hay supplemented with lucerne hay as feed for pregnant goats. The digestibility and metabolisable energy (ME) content of the forages were estimated from the dry-matter (DM) disappearance, which was measured by the rumen-bag technique. Forty-five multiparous Matebele goats (mean liveweight 35.1 kg) in the 14th week of gestation were fed at one of three levels: low (L: 0.26 MJ ME/kg LW^0.75), medium (M: 1.5 x L) and high (H: 2 x L with veld hay offered ad libitum). Maize grain (13.2 MJ ME/kg DM), lucerne hay (7.2 MJ ME/kg DM) and veld hay (5.52 MJ ME/kg DM) were fed at 20, 24 and 56%, respectively, of the total metabolisable energy per treatment. The goats were weighed and body-condition scored (scale 1-10) weekly during pregnancy and within eight hours of parturition.

Results for 42 goats that gave birth to single living kids are presented. Mean ME intakes during pregnancy were 10.23, 5.91 and 4.08 MJ/day (SE = 0.261) for H. M and L groups, respectively. Between week 15 and kidding, changes in net liveweight were +4.8, +0.9 and -1.5 kg (SE = 0.86), and changes in body condition were -0.4, -1.6 and -1.3 (SE = 0.27), for the goats on H. M and L treatments, respectively. Kid birth weights were 3.18, 2.92 and 2.89 kg (SE = 0.154). These birth weights were higher than normally realised in the dry season under natural rearing conditions. It is concluded that Matebele goats can eat sufficient quantities of veld hay to meet energy requirements during pregnancy.

RESUME

Alimentation de chèvres Matebele avec du fourrage de veld au cours des trois derniers mois de gestation

Dans les conditions normales d'élevage, le poids des chevreaux à la naissance et leur vitesse de croissance tombent au plus bas au cours de la saison sèche, au moment où les aliments du bétail sont rares et de très mauvaise qualité. Cette expérience examine l'effet de l'alimentation de chèvres en gestation avec du fourrage de veld complémenté avec du foin de luzerne. La digestibilité des fourrages et leur teneur en énergie métabolisable (EM) ont été estimées à partir du taux de dégradation de la matière sèche, mesuré par la technique des sacs en nylon introduits dans le rumen. Quarante cinq chèvres Matebele pesant en moyenne 35,1 kg et qui se trouvaient dans leur 14^e semaine de gestation ont été divisées en trois groupes. Chaque lot a été soumis à l'un ou l'autre de trois régimes alimentaires caractérisés par leur teneur en énergie, laquelle était faible (F: 0,26 MJ EM/kg PV^0,75), moyenne (M: 1,5 x F) ou élevée (E: 2 x F avec du fourrage de veld ad libitum). Le maïs grain (13,2 MJ EM/kg MS), le foin de luzerne (7,2 MJ EM/kg MS) et le foin de veld (5,52 MJ EM/kg MS) constituaient respectivement 20, 24 et 56% de l'énergie métabolisable contenue dans chaque ration. Les chèvres étaient pesées et leur état d'engraissement enregistré une fois par semaine au cours de la gestation puis dans les huit heures précédant la parturition.

Les résultats présentés ici portent sur les 42 chèvres qui avaient donné naissance à un seul chevreau vivant au cours de cette expérience. L'ingestion moyenne d'EM pendant la gestation était de 10,23; 5,91 et 4,08 MJ/j (erreur type = 0,261) pour les lots soumis respectivement aux niveaux faible, moyen et élevé d'énergie. De la 15^e semaine de gestation à la parturition, les variations de poids vif net étaient respectivement de +4,8; +0,9 et -1,5 kg (erreur type = 0,86) avec des variations correspondantes de l'état d'engraissement égales à -0,4; -1,6 et -1,3 (erreur type = 0,27). Les poids à la naissance des chevreaux étaient de 3,18; 2,92 et 2,89 kg (erreur type = 0,154) respectivement pour les niveaux faible, moyen et élevé d'énergie. Ces valeurs étaient supérieures aux chiffres habituellement enregistrés en saison sèche dans les conditions normales d'élevage. Les chèvres Matebele en gestation peuvent par conséquent consommer suffisamment de foin de veld pour couvrir leurs besoins en énergie.

INTRODUCTION

Goats play a vital role in the livelihood of smallholder farmers in Zimbabwe. Over 70% of the national goat population are found on communal lands within agro-ecological zones IV and V (Vincent and Hack, 1960). As these areas are generally unsuitable for cropping, livestock rearing on natural vegetation is the major agricultural enterprise.

The natural vegetation of these semi-arid areas of Zimbabwe (commonly referred to as sweetveld (Rattray, 1957)) consists mostly of palatable annual grasses which goats willingly graze in the rainy season when they are nutritious (high protein content) and abundant (Nyamangara, 1990). However, during the dry season the grasses are very mature and of low palatability. Moreover, the trees and bushes on which the goats rely for browse shed their leaves. Elliot and Fokkema (1961) showed that the dry-matter intake of range livestock in Zimbabwe during the dry season is unlikely to exceed 1.5%, and may be as low as 1.2%, of body weight. This energy and protein deficiency causes weight loss in all livestock, but the effects are particularly severe in goats. Seventy per cent of the indigenous Matebele goats in the communal areas kid in the dry season; consequently, kid birth weights and growth rates are low (<2 kg and 40 g/day) and kid survival rates below 65% are not uncommon (Ndlovu and Royer, 1988).

Adequate nutrition of goats during pregnancy and lactation can improve kid birth weights and survival rates (Sibanda, 1990). Veld hay, harvested during the rainy season, is a potential low cost feed resource for the dry season, but the responses of pregnant goats to diets based on veld hay are not known. The study reported here examined the effects of three levels of feeding on doe liveweight and body-condition score changes and kid birth weights.

MATERIALS AND METHODS

Location

The experiments were conducted at the former Thuli Breeding Station, Guyu, situated within agro-ecological zone V in Matebeleland Province, south-west Zimbabwe. Total rainfall during the study year (1989) was 362 mm. Mean maximum and minimum temperatures were 28 and 13°C, respectively.

Experimental diet

The experimental diets consisted of maize grain, lucerne hay and veld hay, providing 20, 24 and 56%, respectively, of the total metabolisable energy. Three levels of feeding were investigated: the low level treatment (L) allowed 0.26 MJ ME/kg LW^0.75; the medium level treatment (M) was calculated as 1.5 x L; and the high level treatment (H) was calculated as 2 x L for the purposes of establishing the allowance of maize grain and lucerne, while veld hay was given ad libitum at 1.4 times the previous day's intake.

Feedstuffs

Lucerne hay and veld hay were obtained from Aiselby Municipality Farm, Bulawayo, which is in the same agro-ecological zone as Guyu. The lucerne was harvested at the early flowering stage, left to wilt overnight, baled and then heat-dried in a barn. The veld hay was made from natural annual grasses growing on cleared rangelands during the rainy season. The grass was baled and then dried in the same manner as the lucerne.

Table 1. Chemical composition of maize grain, lucerne hay and veld hay

Feed	Chemical composition (g/kg DM)
	Dry matter (g/kg)	Ash	Crude protein	Neutral detergent fibre
Maize grain	90.0	1.6	8.2	-
Lucerne hay	90.1	9.5	18.1	53.8
Veld hay	91.2	10.8	12.8	72.6

The dry-matter, ash, crude-protein and neutral detergent fibre contents of the feedstuffs were determined and are shown in Table 1.

The rumen degradabilities of the lucerne hay and the veld hay were established using the nylon-bag technique (Mehrez and Ørskov, 1977). Four individually-penned, non-pregnant multiparous indigenous Matebele does fitted with permanent rumen cannulae (40 mm diameter) were used in a 2 x 2 crossover design that investigated the degradability of the two feedstuffs when the does were fed at the low and medium feeding levels. The goats were offered the diets for three weeks before any measurements were made. Water and a salt-mineral lick were freely available at all times.

Representative samples of the two hays were ground through a 3-mm screen using a laboratory hammer mill. Nylon bags containing 4 g of the sample were suspended in the rumen for 3, 12, 24, 36, 48, 60 and 72 hours. After withdrawal, the bags were hand washed in cold water for 30 minutes. A further feed sample was used to determine the amount of soluble material by soaking in water before washing. All the bags were then dried at 60°C to constant weight.

The rumen degradability of the feeds was calculated from the percentage dry-matter loss from the incubated samples, as described by the mathematical model p = a + b(1 - e^-ct) (Ørskov and McDonald, 1979), where p is the percentage dry-matter loss at time t, (a + b) defines the maximum potential degradability (the asymptote) and c is the rate constant of b.

Feeding trial

Forty-five multiparous Matebele goats that had been synchronised for oestrus using progesterone-impregnated sponges, mated after oestrus and diagnosed pregnant by consistently high plasma progesterone concentrations, were individually penned and randomly allocated to one of the three feeding levels from week 14 of gestation. Maize grain and lucerne hay were offered at the same time (0800 hours) and veld hay for the M and L groups was offered in two portions (at approximately 1000 and 1500 hours). Refused forage was collected and a representative sample analysed for crude-protein and neutral detergent fibre contents. Water and a salt-mineral lick were always available.

Does were weighed and body-condition scored (Honhold et al, 1989) at weekly intervals. Kids were weighed within eight hours of birth.

Statistical analysis

Data were analysed by the method of least squares analysis of variance using general linear model procedures (SAS, 1986). Post-kidding weights and body-condition scores were adjusted using the initial weights and scores as covariates

RESULTS

Rumen degradation of lucerne hay and veld hay

The degradability characteristics of the forages are shown in Table 2. Feeding level had no effect on any of the parameters. The lucerne hay contained more soluble material and was degraded at a more rapid rate than the veld hay.

Table 2. Dry-matter degradability constants of lucerne hay and veld hay at low and high feeding levels

Feed	Feeding level	Degradability constants
		a	b	c
Lucerne hay	Low	24.43	40.58	0.0926
Lucerne hay	High	24.05	40.12	0.0901
Veld hay	Low	14.52	56.65	0.0392
Veld hay	High	14.90	54.24	0.0423

From the equation p = a + b(1- e^-ct)

where:

p = dry-matter loss (%)
a = readily degradable fraction (%)
b = potentially degradable fraction (%)
c = rate of degradation of the b fraction
t = time (hoofs)

The effective degradability of organic matter (P) was calculated as from the equation P = a + {bc/(c + k₁)}, where k₁ is the passage rate of digesta from the rumen. As k₁ was not determined in this experiment, P was calculated for a range of k₁ values, as shown in Table 3. The mean P values were then used to provide an estimate of the ME values of the two forages, assuming that P is equivalent to organic matter digested (OMD%). The OMD% was converted to digestible organic matter in the dry matter (DOMD%) using the equation DOMD% = OMD% (100 - ash%)/100 (MAFF, 1984: equation 55). ME was then calculated as ME = 0.15 DOMD% (MAFF, 1984: equation 58). The ME values of lucerne hay and veld hay are shown in Table 3.

Table 3. Derivation of metabolisable energy values for lucerne hay and veld hay from potential organic matter degradability parameters

Feed	Rumen turnover rate (k1)	OMD% a	DOMD% b	ME (MJ/kg DM) c	Mean ME (MJ/kg DM)
Lucerne hay	0.03	56.64	51.26	7.69
	0.04	52.31	47.34	7.10	7.20
	0.05	50.33	45.55	6.83
Veld hay	0.025	43.86	39.13	5.87
	0.03	41.69	37.20	5.58	5.52
	0.04	38.17	34.05	5.11

^a OMD% is effective OM degradability = a + {bc/(c + k₁)}
^b DOMD% = OMD% (100 - ash%)/100 (MAFF, 1984: equation 55)
^c ME = 0.15 DOMD% (MAFF, 1984: equation 58)

Feeding trial

Two does gave birth to twins and a third was pseudo-pregnant. The other 42 does gave birth to single, living kids; results from these does (14 per treatment) are shown in Tables 4 and 5.

The mean crude-protein and neutral detergent fibre contents of the refused veld hay from the H treatment were 125 and 756 g/kg DM, respectively, compared to 128 and 726 g/kg DM for the offered material. Therefore there was little evidence of selective feeding. The H treatment does were able to consume 36 g DM/kg liveweight on average over the period of the experiment (Table 4).

Table 5 shows the liveweight changes and body-condition scores of the does, and the birth weights of the kids. Both M and H treatment does showed a net gain in liveweight over the course of gestation, while the L treatment does lost a small amount of weight. Only H treatment does maintained their body condition. There were no significant treatment effects upon kid birth weight. Although there was a trend for birth weight to decline with energy intake, the differences were small compared to the very large differences in energy intake.

DISCUSSION

The ME values estimated for both hays are lower than the values found by Topps and Oliver (1966). The differences could be attributed to many factors, including the degree of maturity of the forages at cutting and the methods of preservation, as well as the methodologies of establishing the energy values.

Table 4. Least mean feed intakes of goats during late gestation

		High feeding level	Medium feeding level	Low feeding level	SE
Number of animals		14	14	14
Daily dry-matter intake
	Maize grain (g/head)	180	107	74	3.9
	Lucerne hay (g/head)	370	220	150	7.4
	Veld hay (g/head)	930	520	360	34.4
	Total (g/kg)	36	24	16	0.82
Daily ME intake
	MI/bead	10.23	5.91	4.08	0.261
	MJ/kg LW0.75	0.58	0.37	0.26	0.011

Table 5. Least-squares mean doe liveweights and body condition scores and kid birth weights

		High feeding level	Medium feeding level	Low feeding level	SE
Number of animals		14	14	14
Liveweight (kg)
	Week 15	36.4	33.1	35.6	1.45
	Post-kidding	39.8	36.0	33.6	0.92
	Net change	4.8	0.9	-1.5	0.86
Body-condition score a
	Week 15	5.6	5.4	5.4	0.27
	Post-kidding	5.2	3.9	4.2	0.27
	Net change	-0.4b	-1.6	-1.3	0.27
Kid birth weight (kg)		3.18	2.92	2.89	0.154

^a On a scale of 1 (severely emaciated) to 10 (over-fat)
^b Not significantly different from zero

Despite the poor digestibility values of the forages, the goats were able to achieve high levels of intake. The intakes achieved by the H treatment does were well in excess of the 3% of liveweight considered by Devendra and Bums (1983) to be the maximum achievable. These high intakes were achieved without recourse to selection, a characteristic of goats offered poor quality forages (Wahed et al, 1990). The lack of selection could be a reflection of the very poor quality of the veld hay.

The net liveweight gains of H and M treatment does indicate the adequacy of nutrient supply for maintenance and production on these treatments. Does on the L treatment utilised body tissue to support the demands of the foetus, so this treatment had little effect on birth weight. The mean birth weights on all three treatments were greater than those reported by Tawonezvi and Ward (1987) for single kids from Matebele goats under a semi-intensive system on-station.

CONCLUSIONS

Veld hay, even of poor digestibility, supplied as a major part of the diet, can be eaten in sufficient quantities to support the energy needs of Matebele goats in late pregnancy.

ACKNOWLEDGEMENTS

Funding for this work was provided by the International Development Research Centre (IDRC), Canada, and the University of Zimbabwe Research Board.

REFERENCES

Devendra C and Burns M. 1983. Feeding and nutrition. In: Goat production in the tropics. Commonwealth Agricultural Bureaux, Farnham Royal, Slough, UK pp. 90-115.

Elliot R C and Fokkema K 1961. Herbage consumption studies on beef cattle. Rhodesia Agricultural Journal 58:124.

Honhold N. Petit H and Halliwell R W. 1989. Body condition scoring system for goats. Tropical Animal Health and Production 21:121-127.

MAFF (Ministry of Agriculture, Fisheries and Food). 1984. Energy allowance and feeding systems for ruminants. 2nd edition. Chalcombe Publications, Marlow, UK 85 pp.

Mehrez A Z and Ørskov E R. 1977. A study of the artificial fibre bag technique for determining the digestibility of feeds in the rumen. Journal of Agricultural Science, Cambridge 88:645-650.

Ndlovu L R and Royer V. 1988. A comparative study of goat productivity in three different regions of Zimbabwe. In: Harrison J (ed), Proceedings of a workshop on goat development, 11-13 January 1988. French Embassy, Harare, Zimbabwe. pp. 55-61.

Nyamangara M E. 1990. Feeding behaviour chemical composition of diet and feed intake in indigenous goats raised on natural vegetation. MPhil Thesis. University of Zimbabwe, Harare, Zimbabwe. 121 pp.

Ørskov E R and McDonald I. 1979. The estimation of protein degradability in the rumen from incubation measurements weighted according to rate of passage. Journal of Agricultural Science, Cambridge 92:499-503.

Rattray J M. 1957. The grasses and grass associations of associations Rhodesia. Rhodesia Agricultural Journal 54: 197.

SAS (Statistical Analysis Systems Institute Inc). 1986. SAS users' guide. SAS, Cary, North Carolina, USA.

Sibanda R. 1990. Productivity of indigenous goats under an accelerated kidding management system. MPhil Thesis. University of Zimbabwe, Harare, Zimbabwe. 93 pp.

Tawonezvi H P R and Ward H K 1987. Productivity of indigenous sheep and goats. 2. Environmental influences on performance of goats. Zimbabwe Journal of Agricultural Research 25:51-58.

Topps J H and Oliver J. 1966. Animal foods of Central Africa. Rhodesia Agricultural Journal Technical Handbook 2. Government Printer, Salisbury, Rhodesia [now Harare, Zimbabwe]. p. 58.

Vincent V and Hack H R. 1960. Land use and recommended farming systems for different land capability classes in different areas of Southern Rhodesia. Annual Conference of the Professional Officers of the Department of Research and Specialist Services. Government of Southern Rhodesia, Salisbury, Southern Rhodesia [now Harare, Zimbabwe]. pp. 16-27. [ILCA library accession (microfiche) number 56017]

Wahed R A, Owen E, Naate M and Hosking J B. 1990. Feeding straw to small ruminants: effect of amount offered on intake and selection of barley straw by goats and sheep. Animal Production 51:283-289.