S . K. KaruaDepartment of Animal Science
Bunda College of Agriculture
P.O. Box 219
Lilongwe, Malawi
Summary
Resume
Introduction
Materials and methods
Results
Mortality
Seasonal effects
Discussion
Acknowledgements
References
Goat performance in 6 villages around Bunda College and at Lifidzi ranch in Salima was monitored during 1985 and 1986. In the villages age at first kidding was 15.0 ± 1.9 mo: season of the doe's own birth affected her age at first kidding but her type of birth did not. No environmental variables affected age at first kidding on the ranch. Kidding percentage was 107 under village conditions and 175 on the ranch. Kidding interval was 44.9 ± 11.3 wk in the villages and 35.2 ± 6.8 wk on the ranch. Season had a significant effect on kidding interval in the villages but not on the ranch. Village goat litter size was 1.35 ± 0.5; on the ranch it was 1.38 ± 0.5. In both systems litters were largest during the cool dry season (May-Aug) and smallest in the hot dry season (Sep-Nov). Birth weight of single born kids was 1.75 ± 0.4 kg in the villages and 1.77 ± 0.5 kg at the ranch. Birth weight was positively correlated to dam postpartum weight (r = 0.58). Dams kidding multiples were heavier than those kidding singles. Kid mortality was different in different seasons in the villages but not on the ranch. Mortality was 11.0% in the villages and 20.1% on the ranch while abortion was 5.61% in the villages and 9.5% on the ranch.
Les paramètres de production des caprins des 6 villages autour de l'Université de Bunda et au ranch de Lifidzi à Salima furent enregistrés en 1985 et 1986. En milieu villageois l'âge à la première mise bas était de 15,0 à 1,9 mois. La saison de naissance de mère affecte l'âge au premier part mais le type de naissance n'a aucune influence. Aucune des variables lices à l'environnement n'a d'effet sur cet âge au ranching. Le taux de mise bas atteignit 107 p.cent en milieu villageois et 175 p.cent au ranch. L'intervalle de mise bas était de 44,9 ± 11,3 semaines dans les villages et 35,2 ± 6,8 semaines au ranch. La saison influe significativement sur l'intervalle de mise bas dans les villages mais pas au ranch. Dans les 2 systèmes les plus grandes portées sont obtenues pendant la saison sèche fraiche (mai-août) et plus petites pendant la saison sèche chaude (sep-nov). Le poids à la naissance était de 1,75 à 0,4 kg dans les villages et 1,77 à 0,5 kg au ranch. Le poids à la naissance des simples est positivement corrélé avec le poids postpartum des femelles (r = 0,58). Les mères de chevreaux nés multiples étaient plus lourdes que celles des simples. La mortalité des jeunes varie avec les saisons en milieu villageois mats pas au ranch. Le taux de mortalité variait de 11,0 p.cent au village a 20,1 p.cent au ranch alors que le taux d'avortement atteignait 5,6 p.cent dans les troupeaux villageois et 9,5 p.cent sur les troupeaux du ranch.
The goat population of Malawi was estimated at 1.6 million in 1981 (National Sample Survey of Agriculture 1981/1982). Most of the goats are kept in rural areas and are of the Small East African type (Mason and Maule, 1960). A small number of Boer goats is kept on government farms. Local goats are kept primarily for meat. Goat meat is second (17%) to chicken (69%) as the preferred meat, ahead of beef (10%) and mutton (2%) (Khaila and Itimu, in press). As in other parts of Africa, the goat has been neglected by researchers and extension agents in Malawi. Of the 43% of households keeping goats in Lilongwe Rural Development Project area (Central Region), 84% received no extension advice on the goat and 94% received no veterinary services (Khaila and Itimu, 1986). Lack of research has resulted in a scarcity of information that can be used by the extension services.
This study was carried out to provide baseline data. Results obtained could be used as a basis for recommending different management practices for use by the extension services.
Study area
The study was carried out in the Central Region of Malawi at 2 sites. The first was around Bunda College of Agriculture in Lilongwe involving 6 adjacent villages (14°55'S, 33°50'E, 1200 m). The second was at Lifidzi ranch in Salima (13°55'S, 34°28'E, 500 m).
The Central Region has a cool dry season (May-Aug), a hot dry season (Sep-Nov) and a hot wet season (Dec-Apr) which is also the crop growing season. In Lilongwe the soils are ferruginous and vegetation is mostly open canopy woodland. At Salima, soils are alluvial and vegetation is thicket savanna.
Animals and their management
A total of 597 goats (mare: female ratio 1:8) of different ages in the villages and 61 does and 1 buck at the ranch were used in this study. All goats were identified by plastic tags and all village goats were treated as one flock as were those at the ranch.
In the villages, goats were housed in the family house, in the kitchen, in separate purpose built houses ('khola') or in houses occupied by male children. Regardless of housing type, 95% of farmers tied their goats to a post for the night. In Salima, goats were kept in elaborate structures with slatted platforms and a leakproof roof. During the cool and hot dry seasons, goats were left to roam freely and seek their own food during the day in the villages. In the hot wet season they were restricted by tethering (93%) or herding (7%) to prevent crop damage. In Salima, goats were herded from 07:00 to 11:30 and from 13:00 to 17:00 every day and in all seasons.
Data collection
Kiddings, mortalities and abortions were recorded as they occurred. Data collected included kidding dates, litter sizes, sex of kids, birth weight and postpartum weight. No postmortem examinations were carried out on goats which died in Lilongwe. Postmortem was done on all goats which died at Salima but because of lack of dispatching facilities only blood smears were examined by the Central Veterinary Laboratory in Lilongwe.
Kids born during the study were followed to first parturition in order to determine age at first kidding. Age at breeding for the first time was calculated on the basis of a gestation length of 150 d.
Statistical analysis
A one-way analysis of variance was used to analyse the different parameters studied. Means were compared using Duncan's Multiple Range Test. Chisquare was used to compare kid mortality between Lilongwe and Salima. Relationship between dam postpartum weight and kid birth weight was determined using linear regression analysis. All differences were tested at the 5% level.
Breeding performance
A summary of breeding performance is provided in Table 1. Age at first kidding was significantly later on Salima ranch than in the villages. Age at first kidding was not affected by the dam's own birth type under either system.
The kidding percentage (total number of kids born/total number of does exposed in a year) was higher under ranch management (175%) than under village management (107%), the difference not being significant. Kidding interval was significantly shorter under ranch management than under village management. There was no difference in litter sizes between village and ranch managements.
Sex ratios of kids born were in the expected range: of 779 kids born in both systems, 50.8% were male and 49.2% were female.
Weight characters
Overall mean birth weight was not affected by system (ranch and village) or kid sex. Single and twin born kids did not differ in weight at birth but both were heavier than triplets. There were indicators that the expected trends in birth weight (singles v twins, males v females) were present.
Postpartum weight increased with increasing litter size. Dams giving birth to triplets were heavier than those with singles and twins under both management systems. Postpartum weight of does with singles was significantly and positively correlated (r = 0.58) with the birth weight of their kids. There was no correlation between dam postpartum and kid birth weight for multiple litters (r - 0.15).
Table 1. Performance summary of Malawi local goats in Lilongwe villages and at Salima ranch.
|
Parameter |
Lilongwe |
Salima |
||||||
|
n |
|
s.d. |
n |
|
s.d. |
|||
|
Age at first kidding (mo) |
16 |
15.6 |
1.9 |
21 |
17.5 |
3.1 |
||
|
|
Single born |
8 |
15.8 |
1.9 |
8 |
16.6 |
2.0 |
|
|
|
Multiple born |
8 |
15.4 |
2.0 |
13 |
17.5 |
3.2 |
|
|
Age at first conception (mo) |
16 |
10.6 |
1.9 |
21 |
12.5 |
3.1 |
||
|
Kidding percent (%) |
268 |
107 |
60 |
175 |
|
|
||
|
Kidding interval (wk) |
117 |
44.9 |
11.3 |
80 |
35.2 |
6.8 |
||
|
Litter sizes |
422 |
1.35 |
0.5 |
152 |
1.38 |
0.5 |
||
|
Birth weight (kg) |
574 |
1.75 |
0.4 |
209 |
1.77 |
0.5 |
||
|
|
Single |
M |
139 |
2.0 |
0.5 |
45 |
2.0 |
0.6 |
|
|
|
F |
133 |
1.8 |
0.5 |
49 |
1.9 |
0.5 |
|
|
Twin |
M |
92 |
1.7 |
0.5 |
57 |
1.6 |
0.6 |
|
|
|
F |
146 |
1.6 |
0.5 |
55 |
1.6 |
0.6 |
|
|
Triplet |
M |
2 |
1.25 |
0.2 |
3 |
1.5 |
0.0 |
|
|
|
F |
4 |
1.00 |
0,0 |
- |
- |
|
|
Postpartum weight (kg) |
|
|
|
|
|
|
||
|
|
Single |
275 |
27.9 |
6.5 |
96 |
27.8 |
5.0 |
|
|
|
Twin |
144 |
32.4 |
5.2 |
55 |
29.6 |
5.0 |
|
|
|
Triplet |
2 |
36.5 |
3.0 |
1 |
36.5 |
0.0 |
|
|
Mortality (%) |
64 |
11.1 |
42 |
20.1 |
|
|||
|
|
M |
29 |
9.9 |
25 |
23.8 |
|
||
|
|
F |
35 |
12.4 |
17 |
16.3 |
|
||
|
|
Single |
27 |
9.9 |
16 |
17.0 |
|
||
|
|
Twin |
27 |
12.5 |
25 |
22.6 |
|
||
|
|
Triplet |
1 |
16.6 |
1 |
33.3 |
|
||
|
Abortion (% viable births) |
25 |
5.6 |
16 |
9.5 |
|
|||
The mortality rate of kids from 1 d to 150 d was higher under ranch than under village management, without this difference being significant. Since no postmortem examination was conducted in the villages, the cause of death was not established. Postmortem examination and laboratory results from kids which died at the ranch provided only an indication of the major cause of death. Results showed no parasites in 81% of the kids which died, while 11% died of anaplasmosis, 4% of pneumonia and 4% of Haemonchus- induced enteritis. Although more kids from multiple births died than kids born as singles, the difference was not significant. More female (12.4%) than male (9.9%) kids died under village management while under ranch management more male (23.8%) than female (16.3%) kids died.
Most mortalities occurred in kids before 1 mo under both management systems (Table 2). Abortion rates were not significantly different under village (5.6%) and under ranch management (9.5%).
Age at first kidding in the village flocks was lowest for does kidding in the hot wet season although this did not differ from those kidding in the hot dry season (Table 3). The subsequent kidding interval was significantly longer in those dams which kidded in the hot dry season. Fewest does kidded in the cool dry season. Mortality was least in the hot dry season. Although there were some apparent seasonal differences in litter sizes, kid birth weights and abortions in the villages, none of these effects were significant.
Table 2. Age specific mortality of Malawi local goats under two systems of management.
|
Age at death |
Lilongwe traditional villages |
Salima ranch | ||
|
(mo) |
n |
% |
n |
% |
|
< 1 |
22 |
33.8 |
31 |
72.1 |
|
1-2 |
5 |
7.7 |
3 |
7.0 |
|
2-3 |
13 |
20.0 |
4 |
9.3 |
|
3-4 |
11 |
16.9 |
3 |
7.0 |
|
4-5 |
14 |
21.6 |
2 |
4.6 |
On Salima ranch there were no significant effects of season on any of the parameters studied.
Differences in performance between the village and ranch systems appear to be related to management and not to animal factors.
Age at first kidding was significantly later at Salima ranch than in the Lilongwe villages because first mating time was deliberately controlled until goats reached an average age of 12 mot In Lilongwe, under village conditions there was no such control. Age at first kidding obtained in this study can be compared with those reported for Red Sokoto in Nigeria (14.5 mo; Ngere et al, 1984) and Long-legged Sahel in Mali (16.2 mo; Wilson and Durkin, 1988) but appears older than the one reported for Landim goats in Mozambique (12.3 mo; McKinnon and Rocha, 1985).
Table 3. Seasonal effects on performance of Malawi local goats in two management systems.
|
Location and system |
Parameter |
Season |
||||||
|
|
Cool dry (May-Aug) |
Hot dry (Sep-Nov) |
Hot wet (Dec-Apr) |
|||||
|
Lilongwe (village)
|
Age at first kidding (mo) |
|
16.8 |
1.6a |
15.7 |
1.5ab |
13.6 |
0.9b |
|
Kidding interval (wk) |
|
43.8 |
13.7a |
47.8 |
13.1b |
42.4 |
8.4a |
|
|
Kidding percent (per no) |
|
1.5a |
|
14.1b |
|
10.2b |
|
|
|
Litter size (no) |
|
1.52 |
0.5 |
1.37 |
0.5 |
1.31 |
0.5 |
|
|
Birth weight(kg)
|
Single |
2.4 |
0.5 |
2.0 |
0.5 |
1.81 |
0.5 |
|
|
Multiple |
1.9 |
0.5 |
1.8 |
0.6 |
1.7 |
0.5 |
||
|
Mortality (%) |
|
33.3a |
|
8.9b |
|
57.8a |
|
|
|
Abortion (%) |
|
4.2 |
|
33.3 |
|
62.5 |
|
|
|
Salima (ranch)
|
Age at first kidding (mo) |
|
18.0 |
1.9 |
- |
|
17.2 |
3.2 |
|
Kidding interval (wk) |
|
34.7 |
6.7 |
32.1 |
4.2 |
36.4 |
7.8 |
|
|
Kidding percent (per mo) |
|
9.5 |
|
8.1 |
|
7.6 |
|
|
|
Litter size (no) |
|
1.45 |
0 5 |
1.26 |
0.5 |
1.37 |
0.5 |
|
|
Birth weight (kg)
|
Single |
2.1 |
0.6 |
1.8 |
0.6 |
2.0 |
0.5 |
|
|
Multiple |
1.6 |
0.5 |
1.5 |
0.6 |
1.7 |
0.5 |
||
|
Mortality (%) |
|
38.1 |
|
33.3 |
|
28.6 |
|
|
|
Abortion (%) |
|
56.3 |
|
37.5 |
|
6.2 |
|
|
Along rows, values without a suffix letter in common differ (P < 0.05).
The season of birth of the dam affected age at first kidding in the villages but not at the ranch. This is attributed to the way the goats were managed under the traditional system. In the hot wet season goats were tethered both by day and by night which restricted breeding activities. During the dry season goats were loose and mixed freely thus allowing increased breeding activities. Goats reaching breeding age in the wet season therefore usually had to wait until the next dry season for breeding to start.
Management effects also influenced kidding interval and kidding percentage, both of which showed seasonal differences in the villages but not on the ranch. Kidding interval was longer and kidding percentage lower in the villages than on the ranch.
The reasons for bigger litter sizes and heavier birth weights in the cold dry season cannot be clearly established from this study. Feed availability with the onset of the rains in November and December (before tethering starts) and the availability of crop residues in May to July after harvest may, however, contribute to larger litter sizes and birth weights in the cool dry season in the villages. Although the wet season is characterised by an abundance of green forage, this is not available to the goats because of tethering. In conjunction with the increase in internal parasites in the wet season (Chaudhry, 1976; Reynolds, 1979) the limited forage intake due to tethering lowers the performance of the goats. In the wet season, goats in Malawi have lower body weights and are in poorer body condition both in the villages and at the ranch than at other times of year (Karua, in press). The slightly higher percentage of males at birth is as expected (Matthewman, 1977; McDowell and Bove, 1977).
Feed availability and infestation by internal parasites could also contribute to higher mortality and abortion rates in the wet season than in other seasons. Since dams are in poor condition during this period, the ability to maintain pregnancies and produce enough milk for the kids is affected. This is more critical in the villages where there are no attempts to control parasites or to provide supplementary feed. Similar results have been reported for the West African Dwarf goat in Nigeria where mortality rates were higher in kids reared under confinement than in the free roaming or tethering systems (Mack, 1983).
As in other breeds, mortality rates were higher in young animals and those born as part of a multiple birth (Devendra and Burns, 1970; Peacock, 1982; Ngere et al, 1984; Khombe, 1985).
The author is indebted to T P E Makhambera, M W Mfitilodze and J W Banda for their assistance and encouragement in carrying out this research.
Chaudhry, M A. 1976. Fascioliasis and parasitic gastroenteritis. Improvement of Livestock and Dairy Industry - Phase II FAO/UNDP Project MLW/71/519. Food and Agriculture Organisation, Rome, Italy (mimeo).
Devendra D and Burns M. 1970. Goat production in the tropics. Technical Communication No. 19. Commonwealth Bureau of Animal Breeding and Genetics, Commonwealth Agricultural Bureaux, Farnham Royal, UK.
Karua S K. in press. Reproductive performance and growth of the indigenous Malawi goats under traditional system of management. Proceedings of the goat development workshop, January 1988, Bikita, Zimbabwe.
Khaila S W and Itimu O A. in press. Small ruminants and rural development in Malawi: A case study of goat production in Lilongwe Rural Development Project. Proceedings of the workshop on design and implementation of the rural strategy programme, December 1985, Lilongwe, Malawi.
Khombe C T. 1985. Environmental factors affecting the growth and viability of crossbred sheep and goats on range grazing in Zimbabwe. In: R T Wilson and D Bourzat (eds.), Small ruminants in African agriculture. International Livestock Centre for Africa, Addis Ababa, Ethiopia.
McDowell R E and Bove L. 1977. The goat as a producer of meat. Cornell International Agriculture Mimeograph No. 6. Cornell University, Ithaca, USA.
Mack S D. 1983. Evaluation of the productivities of West African Dwarf sheep and goats in southwest Nigeria. Humid Zone Programme Document No. 7. International Livestock Centre for Africa, Ibadan, Nigeria.
Mason I L and Maule J P. 1960. The indigenous livestock of Eastern and Southern Africa. Technical Communication No. 14. Commonwealth Bureau of Animal Breeding and Genetics, Commonwealth Agricultural Bureaux, Farnham Royal, UK.
Matthewman R W. 1977. A survey of small livestock production at the village level in the derived savanna and lowland forest zone of southern Nigeria. M.Sc. Thesis. University of Reading, Reading, UK.
McKinnon S and Rocha A. 1985. Reproduction, mortality and growth of sheep and goats in Mozambique. In: R T Wilson and D Bourzat (eds.), Small ruminants in African agriculture. International Livestock Centre for Africa, Addis Ababa, Ethiopia.
National Sample Survey of Agriculture 1981/1982. Department of National Statistics, Government Printer, Zomba, Malawi.
Ngere L O. Adu I F. Okubanyo I O. 1984. The indigenous goats of Nigeria. Animal Genetic Resources Information 3: 1-9.
Peacock C P. 1982. Seasonal breeding effects on productivity. In: R M Gatenby and J C M Trail (eds.), Small ruminant breed productivity in Africa. International Livestock Centre for Africa, Addis Ababa, Ethiopia.
Reynolds L. 1979. Breeding performance and growth rate of the indigenous Malawi goat. Agricultural Bulletin of Bunda College of Agriculture 10: 90-100.
Wilson R T and Durkin J W. 1988. Livestock production in central Mali: Reproductive components in traditionally managed sheep and goats. Livestock Production Science 19: 523-529.
±
±