P N de Leeuw, P P Semenye, C P Peacock and B E Grandin
7.1 Cattle productivity
7.2 Smallstock productivity
References
7.1.1 Introduction
7.1.2 Herd composition
7.1.3 Breeds and weights
7.1.4 Reproductive performance
7.1.5 Mortality and disease incidence
7.1.6 Growth of young stock
7.1.7 Milk offtake and lactation yield
7.1.8 Productivity index
The major parameters that determine the productivity of a cattle herd are:
· the reproductive performance of the breeding females
· mortality
· growth rates from birth to maturity
· division of milk between calves and people.
Although overall mortality and growth are important determinants of herd performance it is the cow-calf unit that drives the system, in the short-term because of the milk supply and in the long-term because it is the number of calves, their survival and growth that determines the sustained viability of the herd. As a consequence, this study focused on this herd component.
This chapter presents herd composition data by age and sex categories and data on calving rate, calf mortality, calf growth and milk yield and offtake. In the final section these parameters are used in calculating the productivity index of the cow-calf component of the herd.
The structure of 41 herds across the three group ranches was recorded at the beginning of the study (1981-82). In total, over 5000 cattle were classified by age, sex, management category1, breed and weight. The results of the analysis were published by King et al (1984).
1 Management categories were Females: calf, heifer, adult lactating and adult dry; Males: calf, replacement bull; Steers: weaner, immature, mature and large mature.
Table 7.1 shows herd structures for the three wealth classes. All herds had a preponderance of females (65-70%). Larger producers had proportionally fewer females but a larger proportion of immature steers.
There was little difference across ranches in the proportion of cows (35-37%) or of total males (32-34%), although the composition of the latter varied: Mbirikani producers kept a larger proportion of immature steers (10%) than producers on the other ranches (56%). Olkarkar ranch had the largest proportion of mature steers (3.8% vs 0.9% and 1.5% on Merueshi and Mbirikani respectively).
The herds of 41 households were also stratified by weight-for-sex in five herd-size classes (Table 7.2). Herd size had a similar effect on herd composition to that of wealth class, in that the proportion of heavy steers increased with herd size, while there was only a small increase in the proportion of younger, lighter steers. The proportion of bulls in the herd declined with increasing herd size. Since each producer prefers to have his own breeding bulls and replacements, these take up a larger proportion in the smaller herds. King et al (1984) found that the number of cows per bull increased from 11 in poor producers' herds to 14 in herds of rich producers.
Table 7.1. Cattle herd structures by wealth class Olkarkar Merueshi and Mbirikani group ranches, 1981.
|
|
Age |
Per cent of animals by class |
|||
|
Wealth class1 |
Mean |
||||
|
Poor |
Medium |
Rich |
|||
|
Males |
|
|
|
|
|
|
Calves |
0-1 |
8.4 |
10.4 |
6.9 |
7.8 |
|
Young steers |
1-2 |
11.4 |
7.1 |
11.2 |
10.4 |
|
Immature steers |
2-4 |
4.2 |
4.2 |
10.0 |
8.2 |
|
Mature steers |
> 4 |
0.5 |
3.0 |
1.9 |
2.0 |
|
Bulls |
> 4 |
5.7 |
6.3 |
4.9 |
5.3 |
|
Total males |
|
30.2 |
31.0 |
34.9 |
33.7 |
|
Females |
|
|
|
|
|
|
Calves |
0-1 |
10.7 |
10.8 |
9.3 |
9.8 |
|
Heifers |
1-4 |
18.4 |
23.5 |
19.9 |
20.5 |
|
Cows |
> 4 |
40.6 |
34.8 |
35.7 |
36.1 |
|
Total |
|
69.7 |
69.1 |
64.9 |
66.4 |
Columns do not sum to 100 due to rounding.
1 Poor = <5 tropical livestock units (TLU) per active adult male equivalent (AAME); medium = 5-12.99 TLU/AAME; rich = ³ 13 TLU/AAME.Source: Derived from King et al (1984).
Table 7.2. Relationship between herd size and herd composition, Olkarkar, Merueshi and Mbirikani group ranches, 1981.
|
Age/sex class |
Per cent of animals by class |
||||
|
Herd size (head) |
|||||
|
1-40 |
41-80 |
81-150 |
151-300 |
>300 |
|
|
Bulls > 100 kg |
9 |
6 |
8 |
5 |
5 |
|
Steers 100-200 kg |
7 |
6 |
7 |
9 |
8 |
|
Steers >200 kg |
6 |
5 |
11 |
10 |
13 |
|
Total males >100 kg |
22 |
17 |
26 |
24 |
26 |
|
Females 100-200 kg |
9 |
12 |
13 |
16 |
14 |
|
Females >200 kg |
48 |
43 |
43 |
40 |
44 |
|
Total females > 100 kg |
57 |
55 |
56 |
56 |
58 |
|
Ratio: |
0-4 |
0.6 |
0.6 |
0.7 |
0.6 |
|
Ratio: |
0.22 |
0.32 |
0.30 |
0.30 |
0.25 |
|
Per cent of households |
24 |
24 |
23 |
16 |
13 |
|
Per cent of cattle |
4 |
9 |
20 |
23 |
44 |
Large herds (151-300 head) had the smallest proportion of breeding females (defined as those weighing more than 200 kg) but the highest young female/cow ratio and one of the highest calf/cow ratios. The low calf/cow ratio in small (140 head) herds might indicate a lower calving rate in these herds but it is more likely that they were forced to sell or exchange young female stock for cash or marketable steers from the rich and medium-wealth producers. There was little difference between ranches in the proportions of young females and breeding females or in the ratio between these classes.
About 95% of the 5000 cattle included in the weighing exercise were classified as Small East African Zebu; 5% were tentatively classified as mixed-breed (zebu with Sahiwal or Boran). Bulls of mixed blood were commoner on Olkarkar (55% of breeding bulls) and Merueshi (36%) than on Mbirikani, where very few were recorded. Hence the proportion of mixed-blood animals was greatest on Olkarkar. About 19% of calves in the livestock production study were classified as Sahiwal x zebu crossbreds (Semenye, 1987). The percentage of crossbred breeding bulls was higher in herds of poor and medium-wealth producers (23%) than in those of rich producers (15%).
Coat colours of cattle did not differ greatly between ranches, with 70-73% of the cattle having variegated coats. This contrasts with the findings of Finch and Western (1977), that the percentage of light-coloured cattle increased with increasing aridity; they hypothesised that this was because light-coloured animals are better adapted to heat stress and require less water than dark-coloured animals. Dark cattle may be better adapted to low night temperatures and, in view of the altitude (1200 m) of the study area, adaptation to this environmental factor may have been a more important selection criterion than heat tolerance.
Mean weights for the main management classes identified by King et al (1984) are given in Table 7.3. Mean weights of adult females were similar across herd sizes and ranches. As expected, mean steer weight increased with wealth class from 233± 18 kg to 284± 10 kg. Steers were heavier on Olkarkar (311 ±39 kg) than on Merueshi (235±18 kg) or Mbirikani (240±21 kg). Average weight of castrated weaners increased from 141 ± 18 kg on Olkarkar to 208±9 kg on Mbirikani and average weights of female weaners from 140±18 kg to 195±9 kg. There were no differences in weight at weaning between ranches or wealth classes: calves were weaned at 100-120 kg, which corresponds to an average age of 12-14 months, indicating that Maasai prefer long lactation periods (see Section 7.1.7: Milk offtake and lactation yield).
Table 7.3. Mean weights of weaner and adult zebu females, steers and bulls, Olkarkar, Merueshi and Mbirikani group ranches, 1981.
|
Sex |
Mean weight (kg±SE) |
|
|
Weaners |
Adults |
|
|
Female |
174 ± 7a |
251 ± 4a |
|
Steers |
171 ± 7a |
262 ±13a |
|
Bulls |
164 ± 10a |
322 ± 34b |
a Small East African Zebu (SEAZ) only.
b 94 SEAZ' 4 Sahiwal, 14 SEAZ x Sahiwal crossbreds and 2 SEAZ x Boran crossbreds.
Source: Derived from King et al (1984).
Seasonal distribution of birth
The Maasai do not control the breeding of their cattle and hence the reproduction of their cattle is primarily influenced by the bimodal rainfall regime and the resultant seasonality in feed supply. Ideally, carvings should be evenly distributed throughout the year to give a continuous milk supply. In practice, however, there are two major peaks in conceptions that coincide with the two rainy seasons (Figure 7.1). Monthly conception rate was highly correlated with monthly rainfall (r = 0.93). This conception pattern results in a calving peak from the end of the long dry season in
September through November (31% of all births) and a larger peak from February through May (51%). Thus, while over 80% of calves were born during the 8 months when rainfall probability is relatively high, many cows were in the latter half of pregnancy during dry months in either the long or the short dry season.
Calving rate
The average calving rate for the three group ranches was 58%, with Mbirikani showing the lowest (56%) and Merueshi the highest rate (61%). Although the time-span covered by the records was too short to provide long-term estimates of reproductive efficiency of cows, three trends were apparent, relating to:
· the effect of season of birth
· the effect of the length of the milking period
· the high variability in calving intervals.
A total of 196 cows calved during the dry season of 1981; these calved again, on average, 20.8 months later, whereas cows that calved during the rainy period from October 1981 to April 1982 gave birth 16.9 months later. These calving intervals represent calving rates of 58% and 71% respectively. These data suggest that in years with two consecutive good rainy seasons the calving rate could be as high as 75%, whereas if one season's rains failed the calving rate would drop below 60%. Two consecutive poor rainy seasons would reduce calving rate to about 40% (see Section 10.1.2: The herd-projection model).
Figure 7.1. Distribution of cow conceptions between September 1980 and August 1981.
An analysis of records on 144 cows for which both the length of the milking period and the subsequent date of calving was known showed that the duration of milking had little effect on calving interval. When milking was prolonged by one month, the calving interval increased by only 3 days2; cows that were milked for 4 months calved after 20 months and those that were milked for 14 months calved after 21 months. Conception during early lactation was rare: only 7% conceived between 3 and 6 months after parturition. These findings seem to indicate that the stress of pregnancy and early lactation results in anoestrus, the duration of which is almost independent of the length of time over which the cows are milked. Calving intervals were, however, highly variable among the 144 COWS: 43% calved again within 18 months, another 44% between 18 and 24 months after calving and the remaining 13% calved again after 2 years or more (de Leeuw and Wilson, 1988; Semenye, 1987).
2 Wagenaar et al (1986) reported a similar, though more pronounced, effect of milking period on calving interval in pastoral herds in Mali: for every month increase in the milking period, the calving interval was lengthened by 13 days.
Calf survival rates were significantly lower on Mbirikani than on the other two ranches (Table 7.4). Calf survival was high up to 4 months of age due to the efficient management system that Maasai have adopted for young calves which are kept in and around the boma and rely exclusively on their dams' milk (see Section 6.2.4: Calf management). However, mortality during the first few weeks postpartum was poorly recorded and neonatal deaths were not included3.
3 In some pastoral production systems 16% of pregnancies resulted in abortions, stillbirths or neonatal deaths. These causes thus accounted for over a third of all calf deaths up to 1 year old (de Leeuw and Wilson, 1988).
Mortality increased somewhat when calves were sent out to graze, in particular on Mbirikani where only 88% of calves survived to 7 months old. From 7 to 18 months survival was again surprisingly high, being equivalent to a mortality rate of 24% over 11 months (Table 7.4). Calf survival rate was also linked with dam age, calves whose dams were between 5 and 9 years old having the highest survival rates. The main causes of calf death were disease on the northern ranches and disease and malnutrition on Mbirikani (Table 7.5).
Table 7.4. Survival rates of calves to 4, 7 and 18 months on Olkarkar, Merueshi and Mbirikani group ranches, 1981-83.
|
Ranch |
Survival rate at age (months) |
||
|
4 |
7 |
18 |
|
|
Olkarkar |
0.99a |
0.98a |
0.94a |
|
Merueshi |
0.97a |
0.96a |
0.94a |
|
Mbirikani |
0.94b |
0.88b |
0.85b |
|
Mean |
0.97 |
0.94 |
0.91 |
Within columns, numbers followed by the same letter do not differ significantly (P>0.05).Total of 678 calves monitored.
Mortalities in older classes of stock were less systematically recorded but appeared to be mainly due to disease, injuries and predation on the two northern ranches. Mortality rates for cows were lower on Olkarkar and Merueshi than on Mbirikani (2% a year vs 10% a year). Fluctuations in herd mortality due to longer-term variations in forage supply are discussed in Chapter 10 (Section 10.1.2: The herd-projection model).
A general disease survey was carried out from June 1982 to May 1983. Brucellosis and leptospirosis are endemic in the area and were the most common diseases of cattle (Table 7.6). Brucellosis was also the most common disease in goats, whereas anaplasmosis was the most common disease in sheep. The majority of theileriosis cases occurred during an outbreak on Mbirikani following the drought-related movement of cattle to Kuku ranch further south where the main vector for the disease, Rhipicephalus appendiculatus (Brown ear tick), was present. Other diseases reported to be of concern to producers included malignant catarrhal fever, bovine otitis and helminthiasis in calves.
Table 7.5. Causes of calf deaths on Olkarkar, Merueshi and Mbirikani group ranches, 1981-83.
|
Cause |
Percentage of all deaths |
||
|
Olkarkar |
Merueshi |
Mbirikani |
|
|
Disease |
89 |
81 |
51 |
|
Injuries |
7 |
4 |
|
|
Malnutrition |
4 |
|
40 |
|
Predators |
|
4 |
6 |
|
Lost |
|
11 |
3 |
|
Number reported |
27 |
26 |
184 |
Source: Peacock (1984).
Table 7.6. Incidence of major diseases in cattle, sheep and goats in the study area, 1983.
|
Disease |
Disease incidence |
||
|
Cattle |
Sheep |
Goats |
|
|
Brucellosis |
15 |
1 |
7 |
|
Leptospirosis |
18 |
0 |
0 |
|
Paratuberculosis |
2 |
|
|
|
Anaplasmosis |
3 |
4 |
2 |
|
Theileriosis |
4 |
1 |
1 |
|
Babesiosis |
|
1 |
|
|
Bovine otitis |
3 |
|
|
Thus, although several diseases were reported by livestock owners and diagnosed by the veterinary team during this extensive survey, their overall incidence was low. These findings suggest mainly sub-clinical infections and/or enzootic stability and tolerance, indicating low susceptibility to certain diseases and immuno-responsiveness to others. Passive (colostric) immunity provides young stock with their initial resistance to diseases; thereafter young stock build up and maintain immunity by being continuously exposed to the infectious agents. The inherent genetic resistance of the indigenous breeds is believed to play an important role (de Leeuw and ole Pasha, 1987).
The overall mean birth weight of calves was 19.2 kg. Calves born on Olkarkar and Merueshi were significantly (P < 0.05) heavier than those born on Mbirikani (20 kg vs 17.8 kg). Calves were born 2 kg heavier if the last trimester of gestation coincided with a rainy season than if it coincided with a dry period.
Up to the age of 7 months calves on the northern ranches gained weight faster than those on Mbirikani but between 7 and 18 months of age calves on Mbirikani had the higher growth rate (Table 7.7). The differences were, however, not significant.
About 19% of the calves were classed as Sahiwal x zebu crosses, most of which were on Olkarkar. At 4, 7 and 18 months these crosses were 6, 8 and 20 kg heavier than pure zebu animals (P < 0.05)
The effect of season of birth on subsequent growth was significant (P<0.05) only up to the second month. Calves born in the first rains had slightly, but not significantly, higher rates of gain up to 7 months of age than calves born at others times of the year (Table 7.8). The lowest gains were recorded for calves born in the second rains (April-June); their poor performance was due to their entering the long dry season at an early age and their being exposed to poor grazing longer than calves born in other seasons.
Producer wealth class had no significant effect on calf growth rate.
On Olkarkar, calf growth differed significantly (P<0.05) between producers within neighbourhoods, apparently in relation to boma location, which determined the distance to water, watering frequency and range resources available to the calves. Calves from bomas located 5 km from water with adequate grazing between the boma and the water point were 20 kg heavier at 7 months old than calves from bomas 10 km from water with only overgrazed land between the boma and the water point. Variability decreased with age as calves extended their orbit of grazing and relied less on overgrazed areas around the boma and along cattle tracks.
Table 7.7. Daily weight gain and 7- and 18-month weights of calves on Olkarkar, Merueshi and Mbirikani group ranches, 1981-83.
|
|
Number of calves |
Weight gain (g/day) |
Calf weight (kg) at (age): |
|||
|
Calf age (months) |
||||||
|
1-4 |
4-7 |
7-18 |
7 months |
18 months |
||
|
Olkarkar |
140 |
238 |
184 |
199 |
67 |
134 |
|
Merueshi |
143 |
218 |
198 |
204 |
66 |
134 |
|
Mbirikani |
89 |
183 |
179 |
208 |
59 |
129 |
|
Mean |
|
212 |
187 |
204 |
64 |
132 |
Table 7.8. Effect of season of birth on daily weight gain and weight of 7-month-old calves, 1981-83.
|
Season of birth |
Number of calves |
Weight gain (g/day) |
Calf weight (kg) at 7 months old |
|||
|
Calf age (months) |
||||||
|
1-4 |
4-7 |
1-7 |
||||
|
Dry: July-Sept 1981 |
177 |
224 |
188 |
206 |
65 |
|
|
Wet: Oct-Dec 1981 |
98 |
233 |
206 |
213 |
66 |
|
|
Dry: Jan-March 1982 |
48 |
210 |
193 |
208 |
64 |
|
|
Wet: April-June 1982 |
49 |
182 |
162 |
172 |
63 |
|
|
|
Mean |
|
212 |
187 |
200 |
64 |
Source: Adapted from Semenye (1987).
Milk offtake is determined by the interaction of two factors: potential milk offtake from lactating cows and milking strategy. Potential milk offtake was measured by Semenye (1987), who recorded milk offtake from 372 lactating cows once a week in the evening and the following morning. Information on components of milking strategies and their effect on actual milk offtake at the household level was collected subsequently through interviews with women and through re-analysis of the data after including those cows that were milked less often than twice every day (Grandin, 1988).
The availability of milk for consumption in Maasai households is governed by several factors. The potential supply of milk per household depends primarily on herd size, the proportion of lactating cows in the herd and the milk-production potential of each cow. Actual milk supply depends largely on the milking strategy of the producer. This determines how much milk the calf is allowed to suckle and how much is taken off for human consumption. Milking frequency and the amount of milk taken in a milking session are the main components of the milking strategy.
Rich producers milk their cows less often and extract less milk per session than producers studied by Semenye (1987); his yield data should thus be regarded as potential output.
Potential milk offtake
The Maasai have the overall production aim of maintaining a reliable supply of milk to the household throughout the year. This leads to prolonging milking for as long as possible. As the length of the milking period had little effect on the length of the calving interval, the longer the milking period, the greater the milking efficiency of a cow (Table 7.9). However, in a sample of 149 cows Semenye (1987) found that a quarter were milked for less than 6 months, while only 18% were milked for more than 12 months; the overall mean was 9 months. Short lactations were mainly due to the death of the calf and problems with milk let-down.
Table 7.9. Milking period, calving interval and efficiency of milk production.
|
Milking period |
Calving interval |
Efficiency2 |
|
6 |
20.1 |
30 |
|
8 |
20.3 |
39 |
|
10 |
20.5 |
49 |
|
12 |
20.7 |
58 |
|
14 |
20.9 |
67 |
1 Developed from the equation:Y (calving interval) = 19.5 + (0.1 x milking period (months)) (R2 =0.32) (Semenye, 1987:245-248).2 Efficiency = milking period/calving interval.
The average daily milk offtake from cows that were milked twice daily was 0.94 litre. However, offtake varied from 0.65 litre/day in dry months to 1.20 litres/day in wet months. The effect of these differences on milk offtake from the herd was somewhat masked by the seasonality of calving and also by an increase in the proportion of milk taken from cows in early lactation. Milk offtakes given in Table 7.10 represent the means of two dry and two wet seasons, combining the sharp fall in the short dry seasons (February-March) and the much slower but more prolonged decline during the long dry season. The slower decline in milk offtake during the long dry season is mainly related to the relatively large proportion of cows in early lactation following the calving peak from March to May (see Figure 7.1).
Table 7. 10. Effect of season and stage of lactation on daily milk offtake
|
Season |
Milk offtake (litres/cow per day) |
|||
|
Stage of lactation (months) |
Mean |
|||
|
1-3 |
4-6 |
7-9 |
||
|
Rainy seasons1 |
1.16 |
1.13 |
1.02 |
1.09 |
|
Dry seasons2 |
0.92 |
0.76 |
0.73 |
0.79 |
|
Mean |
1.04 |
0.95 |
0.88 |
0.94 |
1 Means of two rainy seasons.
2 Means of two dry seasons.
Source: Semenye (1987).
Lactation yield
Total lactation yield (milk consumed by the calf plus that taken for human consumption) cannot be measured directly under field conditions and must be estimated from calf growth rates together with milk offtake. Daily lactation yield was estimated using growth rates of calves from 30 to 120 days old, during which period growth rate depends mainly on milk intake. Over this period, poor producers on Olkarkar extracted an average of 1.12 litres of milk daily from each milking cow, while calves each gained an average 16.7 kg. This weight gain indicates that each calf consumed approximately 150 litres of milk (Drewry et al, 1959). Thus the total lactation yield over the 90 days was 251 litres or 2.8 litres a day, of which 40% was taken off for human consumption.
Milking strategies and actual milk offtake by wealth class4
4 The following section is based on Grandin (1988) and Grandin (unpublished data). The quantitative information was derived from formal questionnaires administered monthly regarding the number of lactating and milked cattle per sub-household and from fortnightly milk measurements on cows in the animal productivity study (Semenye, 1987). Although the latter data collection was not designed with household consumption in mind, the information can be used to estimate general patterns. Observations combined with informal interviews, mainly in Olkarkar, contributed substantially to the analysis. The available data suggest that patterns in Merueshi were quite similar to those in Olkarkar. Only general statements are possible in relation to Mbirikani because of the drought conditions pertaining on that ranch and the high mobility of both people and stock. There was no information on lactating cattle and what milk records were available were almost exclusively collected from the more accessible bomas.
This section considers the amount of milk taken off for human consumption, which is a function of the potential supply and the needs of suckling calves and the family.
Maasai do not speak of milking cows; they speak of "milking calves". This underscores their understanding of the competition between calves and the family for the milk of the same cow. Maasai know the productive potentials of their animals and their life history. The condition of animals is monitored closely by both the woman who milks them and the head of the household. If a calf seems weak, or becomes ill, its dam will be milked less frequently and the amount of milk taken on each occasion will be reduced. However, Maasai believe that too much suckling can harm a calf; high-yielding cows are milked even if they are temperamental to prevent the calf from consuming too much milk and getting diarrhoea. The amount of milk required by older, grazing calves depends on the availability of forage and water, which was closely related to the season and the location of the homestead. Calves from homesteads near water were taken to water at an earlier age and were watered more frequently than calves from homesteads far from water.
After calf survival, the most important criterion used by a woman in determining how much milk to extract is the need of her family. The amount of milk needed depends on several factors, including the size of the family and its age/sex structure. Women seem to aim for a daily milk offtake of about 1 litre per person in the dry season and 1.5 litres per person in wet season. Seasonal variation in the diet was preferred by most people. However, seasonal variation in milk consumption was a necessity for poor households, whereas for rich households it is by preference.
The availability of other foodstuffs also influenced family needs for milk. In most of the study sites, local shops and markets normally afforded a regular supply of goods and hence the availability of cash governed the supply of other foods. In poor households women milked harder than in rich households, which had more cash available to purchase other foods.
Milk sales accounted for only 5% of milk offtake on Olkarkar and less on Merueshi. Almost no milk was sold on Mbirikani. Demand was highly location specific, with sales made to nearby hotels or to locally resident workers (teachers, game park workers, etc). Thus milk sales did not have a marked effect on milk offtake.
Women did not always milk all their lactating cows. The percentage of cows usually milked generally declined with increasing herd size. Some cows were not milked at all (due to wildness, mastitis, low potential) or were milked for only part of the lactation. Rich households commonly delayed onset of milking and stopped milking earlier in the lactation than did poor households. Thus only some of the lactating cows contributed milk for human consumption at any given time. However, these "usually milked" cows were not necessarily milked every day or at every milking and hence the number of "actually milked" cows was often lower than the number of "usually milked" cows.
Unfortunately, few data are available on the percentage of "usually milked" cows that are actually milked on a given day and estimates were derived from observations and milk recordings. A single data point for households in Olkarkar for July 1982 (mid-long-dry season) indicated that poor households actually milked 95% of their reported "usually milked" cattle, while rich households milked only 70%. The single richest household milked only 58% of the "usually milked" cows. In the very wealthy households, a labour bottleneck at milking limits the number of cows milked; however, this is much less important factor than the need for milk in determining the number of cows milked.
Most households milked their animals twice a day, in the morning and in the evening. The richest households commonly milked their cows only once a day, while others occasionally milked only once a day. The offtake per cow from once-a-day milking was 50-60% that of twice-a-day milking (Semenye, 1987).
Several short-term circumstances commonly resulted in a cow remaining unmilked on one or more occasions. Milking was temporarily suspended if the cow or calf was ill or seemed to be in poor condition. Calves occasionally escaped from the calf-pen and spent the night with their dams, which were consequently not milked in the morning. Calves that were not penned before their dams returned from grazing often met their dams and suckled on the way. Such events were commonest in households with large herds, in which women did not need all the potentially available milk and could afford to be less careful in their calf management. Additionally, women who had more milk available than required took a lot of milk from a few cows rather than taking a little from all their cows, thus reducing the amount of work involved. Women tended to choose animals with younger calves as young calves are easier to handle than older calves and require less milk.
Lastly, actual milk offtake depended on how much milk was taken from each cow milked, which was determined by the number of quarters milked and the degree of stripping. Maasai women usually milked the two left teats, leaving the two right ones for the calf, but milked three quarters when family needs were high. The amount of milk taken from each quarter also varied. The amount of milk given by the cow per unit time decreases after the first few minutes of milking, at which point women with many lactating cows generally moved on to another cow, leaving the rest of the milk for the calf, while poor women coaxed out the last bit of milk.
The effects of wealth class on milking strategies and offtake in Olkarkar are shown in Table 7.11. Milk offtake per person was similar across wealth classes, but the percentage of cows milked, the proportion of cows milked twice a day and the amount of milk taken per cow all decreased with increasing wealth. An offtake of about 1.2 litres per person per day would seem to be the goal in Maasailand, but households with few cows were not able to meet that goal despite a relatively intensive milking strategy. Medium-wealth households met it but with slightly less intensive milking, whereas rich producers achieved this level of offtake using only about one quarter of their potential milk offtake (see Section 10.2.3: Milk offtake).
Table 7.11. Milk-offtake parameters for poor, medium-wealth and rich households on Olkarkar Group Ranch.
|
Parameter/household |
Wealth class1 |
||
|
Poor |
Medium |
Rich |
|
|
Cattle per reference adult |
4 |
7 |
23 |
|
Per cent of lactating cows usually milked |
100 |
70 |
40 |
|
Per cent of lactating cows actually milked2 |
96 |
60 |
30 |
|
Per cent of cows milked twice a day |
88 |
85 |
65 |
|
Daily milk offtake per cow milked (litres) |
1.0 |
0.93 |
0.75 |
|
Total daily milk offtake (litres) |
7.2 |
10.4 |
18.5 |
|
Daily offtake per reference adult (litres) |
1.1 |
1.2 |
1.3 |
|
Actual/potential offtake (%)3 |
86 |
56 |
25 |
1 Poor = <5 tropical livestock units (TLU) per active adult male equivalent (AAME); medium = 5-12.99 TLU/AAME; rich = ³ 13 TLU/AAME;2 Estimated from milk recording observations.
3 The potential is reached when all cows with suckling calves are milked twice a day.
Source: Adapted from Grandin (1988).
Residence and milk offtake in Olkarkar
Neighbourhoods varied markedly in their access to water and the quality and quantity of grazing between the boma and the water point. Frequency of watering of both cattle and calves was inversely related to distance from water, with concomitant effects on both milk production and calves' needs for milk. Most rich producers lived far from the water point to give themselves access to more and better grazing, while most poor producers lived nearer the water point as they had less need for grazing. Watering frequency also varied with neighbourhood. Households 2 km from water watered their stock every day; those about 7 km from water usually watered stock every second day. Distance from water had an effect on milk production and hence on the amount that could be taken for human consumption. Milk yields fell as distance of the boma from water increased, due to lower water intake, longer walking distance to water and reduced grazing times5. Although place of residence was confounded with wealth class, milk offtake was generally lower in households far from water points: e.g. on Olkarkar households 7 km from water had an average milk offtake of 0.78 litres per cow per day, compared with 1.02 litres per cow per day for households 2 km from water.
5 Semenye (1987) has shown that milk offtake on the watering day was about 10% higher than on non-watering days.
Seasonal fluctuations
Daily offtake per cow varied more between seasons than did the number of cows milked (Table 7.12). The number of lactating animals varied between seasons (see Section 7.1.4: Reproductive performance) but variations in the percentage of lactating cows that were usually milked (significant in the case of medium-wealth and rich producers) resulted in smaller seasonal fluctuations in the number of cows usually milked. The percentage of cows that were actually milked seemed to be lower during wet seasons than during dry seasons, particularly in the case of rich producers' herds.
Productivity indices combining cow reproduction, milk offtake per cow and calf viability and growth were used to examine the overall annual output of the cow-calf unit (Table 7.13).
These indices indicated productivity of 53-73 kg of calf/cow per year, or 21 -28 kg of calf/100 kg of cow liveweight per year. This is somewhat higher than in other traditional production systems in similar environments in sub-Saharan Africa, in which indices range from 17 to 23 kg of calf/100 kg of cow (de Leeuw and Wilson, 1988). The productivity of Mbirikani was some 25% less than that of the two northern ranches, mainly because of a minor drought in 1982.
Although these indices provide useful overall yardsticks to measure system productivity, caution is needed in interpreting them because of possible differences in productivity between wealth classes. The effect of wealth class on the productivity indices was thus calculated for Olkarkar. Since there was no evidence that cow and calf survival or calving percentage differed between producer groups, it follows that only calf growth and milk offtake yield influenced the productivity index (Table 7.14). Calves in medium-sized herds were heavier at one year old than those in large or small herds and medium-sized herds had the highest productivity index. Large herds had the second highest productivity index when this was calculated using potential milk offtake but the lowest index when actual milk offtake (derived from Table 7.11) was used in the calculation. This is because rich producers used only about 25% of their potential milk offtake during the favourable conditions of the study period. The contribution of milk offtake to the productivity index is rather small as a result of converting milk offtake to a calf-growth equivalent. This does not reflect the true importance of milk in Maasai households.
Productivity varied much more between individual cows than it did between herds. The major differences were in calving rate and milk yield. In addition, "gift cows" of unknown parity had higher calf mortality and produced calves that weighed less at 12 months old than did cows in their fourth or fifth parity. Combining these differences in production parameters indicates that the productivity of a good cow may be 56% higher than that of a poor cow (Table 7.15).
Table 7.12. Estimates of daily milk offtake in poor, medium-wealth and rich households by season, December 1981- February 1983, Olkarkar Group Ranch.
|
Season |
Wet |
Dry |
Wet |
Dry |
Wet |
|
Period |
Dec-Jan |
Feb-Mar |
Apr-May |
June-Oct |
Nov-Feb |
|
1981/82 |
1982 |
1982 |
1982 |
1982/83 |
|
|
Poor1 |
|
|
|
|
|
|
Daily offtake (litres/cow) |
1.28 |
0.66 |
1.26 |
0.93 |
1.19 |
|
Cows usually milked |
4.9 |
6.3 |
7.8 |
7.6 |
7.1 |
|
Per cent of cows actually milked2 |
96 |
98 |
90 |
96 |
95 |
|
Actual offtake (litres/household per day) |
6.0 |
4.1 |
8.8 |
6.8 |
8.0 |
|
Medium-wealth1 |
|
|
|
|
|
|
Daily offtake (litres/cow) |
1.04 |
0.73 |
0.92 |
0.72 |
1.14 |
|
Cows usually milked |
10.2 |
11.6 |
12.5 |
12.2 |
13.2 |
|
Per cent of cows actually milked2 |
80 |
86 |
80 |
92 |
75 |
|
Actual offtake (litres/household per day) |
8.4 |
7.1 |
9.1 |
8.0 |
11.3 |
|
Rich1 |
|
|
|
|
|
|
Daily offtake (litres/cow) |
0.68 |
0.51 |
0.73 |
0.60 |
079 |
|
Cows usually milked |
24.7 |
24.9 |
28.9 |
23.8 |
23.9 |
|
Percent of cows actually milked |
65 |
80 |
60 |
72 |
67 |
|
Actual offtake (litres/household per day) |
1.0.8 |
10.2 |
12.7 |
10.2 |
12.7 |
1 Poor = < 5 tropical livestock units (TLU) per active adult male equivalent (AAME); medium = 5-12.99 TLU/AAME; rich = ³ 13 TLU/AAME;
2 Estimated from observations and milk recordings.
Table 7.13. Productivity parameters and productivity indices for cattle on Olkarkar, Merueshi and Mbirikani group ranches 1.
|
Parameter |
Ranch |
Overall |
||
|
Olkarkar |
Merueshi |
Mbirikani |
||
|
Cow survival (%) |
98 |
98 |
90 |
95 |
|
Calving percentage |
57 |
61 |
56 |
58 |
|
Calf survival (%) |
95 |
93 |
87 |
92 |
|
Calf weight at 1 year (kg) |
98 |
97 |
91 |
95 |
|
Milk offtake (kg/lactating cow per year) |
250 |
294 |
227 |
257 |
|
Average cow weight (kg) |
240 |
260 |
253 |
251 |
|
Productivity indices |
|
|
|
|
|
kg calf/cow per year1 |
68 |
73 |
53 |
65 |
|
kg calf/100 kg cow liveweight |
27 |
28 |
21 |
25 |
1 The index was calculated as: (cow viability x calving rate x calf survival x calf weight at 1 year (kg)) + (cow viability x calving rate x (milk offtake (kg)/9))
Finally, it must be stressed that these calculations were based on data from only 18 months. Long-term herd productivity is discussed in Chapter 10, in which the productivity index is extended to indicate the productivity of the whole herd, rather than just the cow-calf component.
Table 7.14. Productivity parameters and productivity indices for poor, medium-wealth and rich producers, Olkarkar Group Ranch.
|
Parameter |
Wealth class1 |
||
|
Poor |
Medium |
Rich |
|
|
Calf weight at 1 year (kg) |
89 |
108 |
102 |
|
Potential milk offtake (kg/cow per year) |
290 |
275 |
260 |
|
Index: kg calf/100 kg cow |
27 |
31 |
29 |
|
Actual milk offtake (kg/cow per year) |
250 |
154 |
65 |
|
Index: kg calf/100 kg cow |
26 |
28 |
24 |
1 Poor = <5 tropical livestock units (TLU) per active adult male equivalent (AAME); medium = 5-12.99 TLU/AAME; rich = ³ 13 TLU/AAME.
Table 7.15. Minimum and maximum cow productivity parameters and resultant indices.
|
|
Minimum |
Maximum |
|
Cow survival (%) |
98 |
98 |
|
Calving percentage |
58 |
73 |
|
Calf survival (%) |
86 |
93 |
|
Milk offtake kg/cow per year |
88 |
99 |
|
Calf weight at 1 year (kg) |
225 |
290 |
|
Index (kg calf/cow) |
57 |
89 |
7.2.1 Introduction
7.2.2 Flock composition
7.2.3 Reproductive performance
7.2.4 Mortality and disease incidence
7.2.5 Growth performance
7.2.6 Productivity index
This section focuses on the composition of sheep and goat flocks and their reproductive performance, mortality and growth. It does not consider other components of research on smallstock, such as the relationships between productivity, flock management, and rangeland resource utilisation. Some of these research topics have been reported in de Leeuw and Peacock (1982) and Peacock (1984) and were summarised in Chapters 5 (The study area: Socio-spatial organisation and land use) and 6 (Labour and livestock management). The socio-economic aspects of keeping smallstock are dealt with in Chapters 8 (Livestock transactions, food consumption and household budgets) and 9 (An economic analysis of Maasai livestock production).
Flock structures were determined using the same households as those for cattle herds (see Section 7.1.2: Herd composition). In total, some 2700 sheep and 2300 goats in 41 households were counted and classed according to sex, age and breed (King et al, 1984).
Sheep
The average composition of sheep flocks is given in Table 7.16. There were no significant differences between wealth classes or ranches in the proportion of females in the flocks, which averaged 67%. However, while the distribution of females among age classes was similar on Olkarkar and Merueshi, on Mbirikani over half the females were more than 30 months old (Figure 7.2).
Table 7.16. Average sheep flock structure, Olkarkar Merueshi and Mbirikani group ranches, 1981.
|
Age (months) |
Percentage of flock by class |
||
|
Males |
Castrates |
Females |
|
|
Young (0-15) |
8 |
10 |
21 |
|
Mature (15-30) |
2 |
5 |
20 |
|
Old (>30) |
1 |
6 |
26 |
|
Total |
11 |
21 |
67 |
Derived from King et al (1984).
The proportion of castrates decreased slightly from north to south (24 vs 20%), while rich households retained a larger proportion of castrates of more that 30 months old than did poor households (13 vs 8%) (Figure 7.3), indicating that poor producers sold male stock at an earlier age than rich producers. Olkarkar had the smallest proportion of young males and the highest proportion of young castrates, indicating the producers on this ranch castrated male sheep at an earlier age than did those on the other two ranches (Figure 7.3; see Section 8.2: Livestock utilisation: Transactions for offtake and acquisition). There was an average of 14 ewes per breeding ram, ranging from 12 on Mbirikani to 19 on Olkarkar, and from 11 in poor households to 16 in rich households.
Goats
The number of females and the age distribution in goat flocks was similar to that in sheep (Table 7.17). As with sheep, more than half the female goats on Mbirikani were 30 months old or older (Figure 7.4). The proportion of castrated males was similar on all ranches but old castrates accounted for half of all castrates on Mbirikani, compared with 16% on Olkarkar and 12% on Merueshi (Figure 7.4). The proportion of old castrates also increased with increasing household wealth, from less than 4% in poor households to almost 10% in rich households (Figure 7.5). The mean number of does per breeding buck was 26, ranging from 24 on Merueshi to 30 on Mbirikani and from 13 in poor households to 40 in rich households.
Males and castrates comprised more than 32% of the total flock in the study area, compared with only 5% for the Afar in Ethiopia, 23% for the Daju and the Baggara in the Sudan, 25% for the Bambara and 27% for the Fulani in Mali (Wilson, 1982; Peacock, 1984). In addition to the high proportion of males and castrates in the flocks, 51% of the castrated goats and 35% of the castrated sheep were over the optimum sale age (Peacock, 1984). Sales of small ruminants, especially by rich producers, can thus be doubled without impairing the reproductive capacity of the breeding flock (see Section 8.2: Livestock utilisation: Transactions for offtake and acquisition).
Table 7.17. Average goat flock structure, Olkarkar, Merueshi and Mbirikani group ranches, 1981.
|
Age (months) |
Percentage of flock by class |
||
|
Males |
Castrates |
Females |
|
|
Young (0-15) |
7 |
9 |
18 |
|
Mature (15-30) |
1 |
7 |
22 |
|
Old (>30) |
1 |
8 |
27 |
|
Total |
9 |
24 |
67 |
Derived from King et al (1984).
Breeds
The major sheep breeds were Red Maasai, Black-headed Somali and some Dorpers and their crosses. The fat-tailed Maasai sheep was the predominant breed on the northern ranches (65-75%), while the fat-rumped Somali was the commonest breed on Mbirikani (65%). King et al (1984) found that Dorpers accounted for 20% of the sheep on Olkarkar and 8% on Merueshi, whereas Peacock (1984) stated that only a few Dorpers were observed in some richer Olkarkar households. Almost all the goats were of the Small East African breed.
The Maasai try to control breeding of their smallstock using breeding aprons and this results in a distinct peak of conception early in the long dry season, when the breeding apron was normally removed. However, lambing and kidding occurred throughout the year, albeit with 80% of births taking place between October and April (Figure 7.6), coinciding with the two rainy seasons.
Over the 2-year study period two-thirds of all births on Mbirikani occurred in the first year (1981/82). Lambing and kidding rates were low as the result of low and poorly distributed rainfall between June 1981 and November 1982 and a severe outbreak of Nairobi Sheep Disease in 1982/83 (see Section 7.2.4: Mortality and disease incidence). Between June 1981 and June 1983 only 24% of the sheep and 17% of the goats gave birth twice. These had mean parturition intervals of 12.3 months and 13.6 months respectively. This poor reproductive performance was confirmed by rapid surveys on Mbirikani between 1981 and 1984: 36% of the potential breeding females had not conceived at all; of those that did conceive, some 50-70% did so within 18 months, whereas another 20-25% had a parturition interval of over 2 years (Figure 7.7).
The effect of nutrition during the mating season, particularly on goats, was demonstrated by differences in mating and subsequent birth rates in smallstock flocks on Mbirikani, some of which were moved to Acacia tortilis woodlands south of the ranch to feed on acacia pods during the long dry season in July-August. Comparison of the reproductive performance of flocks that remained on the group ranch and those that moved showed a near-five-fold increase in the percentage of goats that were mated and hence a six-fold increase in the percentage giving birth (Table 7.18). Pod feeding had less effect on sheep reproductive performance.
Mortality rate up to weaning was lower for sheep (18%) than for goats (34%), although the difference was smaller at 18 months (57% vs 66%). The high pre-weaning mortality rate in goats was due in part to their larger litter size; about 15% of the goats produced twins, which were twice as likely to die before weaning as were single-born kids. Only 1% of sheep gave birth to twins. Preweaning mortality rates differed little between ranches but mortality rates from 5 to 18 months and from 0 to 18 months were markedly lower on Merueshi than on the other two ranches (Table 7.19). Mortality rates of goats also differed substantially between wealth classes (Table 7.20); apparently, households with many cattle took less care of their goats than did households with few cattle. Season of birth affected pre-weaning mortality rate in sheep but not in goats; lambs born in the long dry season had higher death rates than those born in other seasons (Table 7.19). Browse was a more important source of feed for goats than for sheep, and this was the most likely cause of the lower dry-season mortality of unweaned kids (de Leeuw and Chara, 1985).
Table 7. 18. Effect of feeding on acacia pods in 1983 on the reproductive performance of goats and sheep, Mbirikani Group Ranch.
|
|
Reproductive performance |
|||
|
Goats |
Sheep |
|||
|
Pods |
No pods |
Pods |
No pods |
|
|
Mated |
97 |
20 |
73 |
47 |
|
Conceived |
80 |
20 |
54 |
47 |
|
Birth |
79 |
13 |
54 |
44 |
|
Abortion |
1 |
7 |
0 |
13 |
Source: Adapted from Peacock (1984), Table 5.4.2., page 245. See also de Leeuw et al (1986).
Table 7.20 shows the causes of death of young (suckling) and adult sheep and goats between August 1981 and February 1983, based on monthly interviews with producers. Disease was a major cause of pre-weaning death in both species and on all ranches. Predators accounted for a large proportion of deaths among young sheep and goats on Olkarkar and of young sheep on Merueshi, but were of little importance on Mbirikani.
The distribution of sheep mortality rates among households was uneven; on all ranches, 60% of the households had low mortality rates (0-10%), while another 7% had rates exceeding 50%, often on account of Nairobi Sheep Disease. The distribution of goat mortality rates was more even; 25% of households had mortality rates of less than 10%, whereas in another 25% death rates were over 60%.
Table 7.19. Mortality rates of smallstock by ranch, wealth class and season of birth.
|
|
Mortality rate (%) |
|||||
|
Sheep |
Goats |
|||||
|
Age (months) |
Age (months) |
|||||
|
0-5 |
5-18 |
0-18 |
0-5 |
5-18 |
0-18 |
|
|
Ranch |
|
|
|
|
|
|
|
Olkarkar |
10 |
26 |
36 |
25 |
36 |
61 |
|
Merueshi |
8 |
15 |
23 |
29 |
18 |
47 |
|
Mbirikani |
10 |
34 |
44 |
32 |
35 |
67 |
|
Wealth class1 |
|
|
|
|
|
|
|
Poor |
10 |
20 |
30 |
9 |
23 |
32 |
|
Medium |
7 |
13 |
20 |
23 |
17 |
40 |
|
Rich |
7 |
14 |
21 |
40 |
13 |
53 |
|
Season of birth |
|
|
|
|
|
|
|
Oct-Dec |
8 |
20 |
28 |
30 |
27 |
57 |
|
Jan-Mar |
10 |
30 |
40 |
29 |
39 |
68 |
|
July-Sept |
16 |
15 |
31 |
29 |
18 |
47 |
1 Poor = <5 tropical livestock units (TLU) per active adult male equivalent (AAME); medium = 5- 12.99 TLU/AAME; rich = ³ 13 TLU/AAME.Source: Adapted from Peacock (1984).
Lambs and particularly kids suffered from scouring, often leading to dehydration, emaciation and death. Scouring was associated with coccidiosis, enterotoxaemia and enteric coli-bacillosis. Another likely cause was salmonellosis. Helminthiasis and coccidiosis were diagnosed frequently in smallstock. Strongyle eggs were found in 30% of faeces samples, and coccidial oocysts in 20%, during the general disease survey, while less than 2% of the animals examined had tapeworm and liver fluke. Enterotoxaemia was identified by post-mortem examination in three separate flocks in Mbirikani, in one of which 80% kid mortality was recorded. Pneumonia caused by Pasteurella haemolytica was also identified as a possible cause of death in lambs.
Tick-borne diseases, including theileriosis, babesiosis, Nairobi Sheep Disease, heart-water and anaplasmosis, were a major cause of adult mortality. However, three-quarters of all smallstock examined had low tick burdens. Anaplasma was the most common blood parasite in both sheep and goats. Babesia were commoner in sheep, and Theileria parasites in both sheep and goats, on Olkarkar and Merueshi than on Mbirikani because of the greater incidence of ticks in the two northern ranches. Other causes of adult mortality were pregnancy toxaemia, particularly during the long dry season, and acute haemonchosis, which was commonest in goats. The study area has, in the past, suffered epidemics of Contagious Caprine Pleuro-Pneumonia but occurrence has been irregular and the last outbreak was reported in 1978.
The most important disease that affected adult sheep and goats during the course of the study was Nairobi Sheep Disease. The first outbreak occurred on Mbirikani in January 1983 and eventually subsided in June 1983; it also spread northwards into Merueshi and Olkarkar. Mortality and abortion rates were high, which, combined with the poor grazing conditions in Mbirikani during the 1982 mating period, caused extremely poor reproduction. In light of its large impact on the sheep and goat flocks in the area, a brief description of the course of the disease is given below.
The failure of the long rains in 1982 on Mbirikani caused households to move cattle, sheep and goats off the ranch; most households moved south into Kuku Group Ranch (see Section 5.3.3: Grazing patterns and stocking rates in the southern ranch). This ranch is on the edge of an area where Nairobi Sheep Disease is enzootic, centred on the foothills of Mount Kilimanjaro (Davies, 1978). There were no working dips in Kuku and most households had not taken their hand-spraying pumps or supplies of acaricide with them during their extensive migration.
Most households returned to Mbirikani following the good rains in November and December 1982. By January 1983 there were reports of a mysterious disease that was killing adult sheep and, to a lesser extent, goats. The outbreak was at its most severe during February and March and subsided by June 1983. Some 57% of sample households were affected. Mortality rates ranged from 16% to 100% in both sheep and goats, with a mean of 44% in sheep and 41 % in goats. In three flocks, only sheep were affected. Some 30% of animals infected recovered. Most Maasai said that there were more abortions during that year than in other years, although the abortion rate (approximately 5-10%) was lower than might have been expected.
Lambs and kids
Growth rates differed markedly between species. Kids grew much more slowly than lambs up to 5 months old, in part because of the higher twinning rate of goats (Table 7.21). Single-born animals were heavier at birth and up to 5 months old than twins. The difference narrowed on 1-year-old animals as a result of high mortality among twins; surviving twins were usually the heavier animals.
Season of birth had a marked effect on subsequent growth rate. Lambs and kids born in the first rains were heavier up to 5 months old than those born in other seasons. Between November 1982 and February 1983, 8- to 12-month-old kids gained an average of 50 g/day, compared with a mean of 25 g/day in other seasons.
Growth rates of both sheep and goats were generally lower on Olkarkar than on the other two ranches (Table 7.21). This may have been related to the higher disease risk, less effective management, generally higher stocking rates and lower availability of browse on Olkarkar (see Chapter 4: The study area: Biophysical environment, and de Leeuw and Chara (1985)). The relatively high postweaning weights of lambs on Mbirikani may have been due, in part, to the high proportion of Black-headed Somali sheep on this ranch.
Table 7.21. Least squares mean weights of lambs and kids at birth and 3, 5, 12 and 18 months old on Olkarkar, Merueshi and Mbirikani group ranches.
|
|
Liveweight (kg) at age (months) |
|||||
|
0 |
3 |
5 |
12 |
18 |
||
|
Lambs |
|
|
|
|
|
|
|
Ranch |
|
|
|
|
|
|
|
|
Olkarkar |
3.4 |
9.4 |
13.0 |
18.8 |
27.5 |
|
|
Merueshi |
3.1 |
10.3 |
13.9 |
20.5 |
28.5 |
|
|
Mbirikani |
4.0 |
10.3 |
14.6 |
23.4 |
30.6 |
|
Overall mean |
3.5 |
10.0 |
13.8 |
20.9 |
28.8 |
|
|
Kids |
|
|
|
|
|
|
|
Birth type |
|
|
|
|
|
|
|
|
Single |
3.4 |
8.7 |
11.3 |
18.7 |
24.4 |
|
|
Twins |
2.7 |
7.1 |
9.4 |
17.4 |
23.9 |
|
Ranch |
|
|
|
|
|
|
|
|
Olkarkar |
3.1 |
7.7 |
9.7 |
15.5 |
19.4 |
|
|
Merueshi |
2.9 |
7.9 |
11.0 |
20.0 |
26.5 |
|
|
Mbirikani |
3.2 |
8.2 |
10.3 |
18.6 |
26.5 |
|
Overall mean |
3.1 |
7.9 |
10.3 |
18.0 |
24.1 |
|
Source: Peacock (1984).
Adults
Weight changes of adult males and females were monitored in three Mbirikani flocks from April 1982 to June 1983. These flocks thus passed through the long 1982 dry season and the excellent rains from late October to January 1983. Final weights coincided with the end of the very poor rains in April and May 1983.
In general, rams maintained their weight through the 1982 dry season, whereas bucks made small but steady gains. In October 1982, at the beginning of the rains, rams weighed an average of 34 kg while bucks weighed 40 kg. At the end of the rains (January 1983) rams weighed 40 kg and bucks weighed 47 kg. Both rams and bucks then maintained their weights until June 1983. Thus males had an average annual growth rate of about 18 g/day (67 kg/year). Similar trends were found in females; their weight remained constant at 32 kg during the long dry season, rose sharply during the rains, partly as a result of pregnancy, to 37-38 kg, and then remained steady until June 1983. Their annual weight gain was thus slightly less than that of males at 5-6 kg. However, weight-changes of breeding females during the dry season were also influenced by the selection of the dry season area. Ewes and does that were taken to the Acacia tortilis woodlands in the south were 6 kg and 4.5 kg heavier respectively than those that remained at the ranch.
Post-partum weights of ewes and does averaged 28 kg, ranging from 25 kg in young animals to about 30 kg in old animals. Effects of breeding season and ranch were significant but small. Both sheep and goats were heavier on Mbirikani (2.0 and 0.7 kg respectively) than on the other two ranches due to a preponderance of older animals in the Mbirikani flocks Dams that dropped offspring in January-February after the first rains were 2.2-2.5 kg heavier than those that gave birth earlier.
The overall productivity of sheep and goat flocks was low, ranging from only 29 g of weaned weight/kg of flock biomass in goat flocks on Mbirikani to 107 g/kg in sheep flocks on the northern ranches (Table 7.22). The productivity of sheep was generally higher than that of goats because sheep had lower pre-weaning mortality rates and lambs weighed more than kids at 5 months and 18 months. Smallstock on Mbirikani were less productive than those on the northern ranches, mainly because of their low reproductive rate during the minor drought in the second year of the study. Output per kg of flock biomass was depressed by the relatively large proportions of castrates in the flocks. Output per kg of breeding female was depressed by the many infertile females in the flocks.
Table 7.22. Productivity parameters and productivity indices for sheep and goat flocks on the northern ranches (Olkarkar and Merueshi) and Mbirikani.
|
Parameter |
Northern ranches |
Mbirikani |
|||
|
Sheep |
Goats |
Sheep |
Goats |
||
|
Births per breeding female |
0.48 |
0.53 |
0.27 |
0.16 |
|
|
Litter size |
1.01 |
1.29 |
1.01 |
1.34 |
|
|
Survival to weaning |
0.90 |
0.75 |
0.90 |
0.68 |
|
|
Survival to 18 months |
0.64 |
0.39 |
0.55 |
0.33 |
|
|
Weight at weaning (kg) |
13.0 |
9.7 |
14.6 |
10.3 |
|
|
Weight at 18 months (kg) |
27.5 |
19.4 |
30.6 |
26.5 |
|
|
Productivity indices |
|
|
|
|
|
|
g/kg biomass of flock:1 |
|
|
|
|
|
|
|
at weaning |
107 |
98 |
60 |
29 |
|
|
at 18 months |
159 |
102 |
77 |
34 |
|
g/kg biomass of breeding females:1 |
|
|
|
|
|
|
|
at weaning |
201 |
172 |
110 |
52 |
|
|
at 18 months |
299 |
179 |
150 |
61 |
1 Number/biomass of old, mature and 50% of young females.
Source: Peacock (1984).
At first sight it may appear that the restriction of the breeding season to 3-4 months in the long dry season may have been a major cause of the poor reproductive performance of smallstock in the study area. It can be argued that breeding stock were in poor condition during the mating season because poor second rains in 1982 and 1983 (March-May) prevented recovery of dams following the previous breeding season. However, although Maasai attempt to restrict breeding to the long dry season, distribution of birth and parturition intervals indicate that control is only partial. At least 20% of the young were born out of season (April-September) and 40-50% of the females that did give birth had intervals of 12-18 months. Nevertheless, although not entirely effective, restriction of the breeding period seems to contribute to the poor reproductive rate in years of below-average rainfall.
As is shown in Chapter 10, the probability of failure in the long rains was high: some 55% of rainy seasons in Olkarkar lasted less than 1.5 months. The probability of poor rains increased with decreasing rainfall from north to south. Some Maasai, particularly those in Mbirikani, countered this risk by moving their flocks to areas that either were rich in browse species or had pod-bearing acacia trees. If good rains or mobility ensure high conception, then the period during which the Maasai mate their smallstock is ideal; young born during the short rains have the longest possible period of good grazing, which leads to high survival rates and good growth. Research in a semi-arid area in Isiolo District in north-east Kenya showed that the productivity of goats was highest when good grazing was available from birth to weaning, provided conception rates were high (Schwartz and Said, 1987).
Limiting the period of breeding has merit in that it produces economies of scale when guarding lambs and kids staying around the homestead and when matching dams and young for suckling in the morning and evening. This work is mainly done by women and children. If breeding was year-round these tasks would go on continuously without respite, preventing women from performing other urgent task (see Chapter 6: Labour and livestock management).
Davies F G. 1978. Nairobi Sheep Disease in Kenya. The isolation of virus from sheep and goats, ticks and possible maintenance hosts. Journal of Hygiene 81:259-265.
Drewry K J. Brown D J and Honea R S. 1959. Relationships among factors associated with mothering ability in beef cattle. Journal of Agricultural Science 18:938-946.
Finch V A and Western D. 1977. Cattle colours in pastoral herds: Natural selection or social preference? Ecology 58:1384-1392.
Grandin B E. 1988. Wealth and pastoral dairy production. A case study from Maasailand. Human Ecology: An Interdisciplinary Journal 16:1-21.
King J M, Sayers A R. Peacock C P and Kontrohr E. 1984. Maasai herd and flock structure in relation to livestock wealth, climate and development. Agricultural Systems 13:21-56.
de Leeuw P N and Chara P. 1985. Species preferences in browse of sheep and goats in mixed Maasai flocks in eastern Kajiado, Kenya. In: Proceedings of the Fourth Small Ruminant CRSP Workshop, Kakamega, March 1985. Small Ruminant Collaborative Research Support Program, Nairobi, Kenya. pp. 252-261.
de Leeuw P N and ole Pasha I. 1987. The cost of tick control in Maasai herds in eastern Kajiado District, Kenya. In: Proceedings of the Sixth Small Ruminant CRSP Annual Workshop, Nairobi, November 1987. Small Ruminant Collaborative Research Support Program, Nairobi, Kenya. pp. 171-180.
de Leeuw P N and Peacock C P. 1982. The productivity of small ruminants in the Maasai pastoral system. In: Proceedings of the First Small Ruminant CRSP Workshop. Small Ruminant Collaborative Research Support Program, Nairobi, Kenya. pp. 118-138.
de Leeuw P N and Wilson R T. 1988. A comparison of cattle productivity in five traditional livestock systems in sub-Saharan Africa. Quarterly Journal of International Agriculture 26:377-390.
de Leeuw P N. Peacock C and Cisse M. 1986. The importance of Acacia tortilis pod feeding in smallstock management in Mbirikani Group Ranch, Kajiado District. In: Proceedings of the Fifth Small Ruminant CRSP Workshop, Nairobi, 4-6 November 1986. Small Ruminant Collaborative Research Support Program, Nairobi, Kenya. pp. 233-241.
Peacock C P. 1984. The productivity of small stock in three group ranches in Kajiado district, Kenya. PhD thesis, Reading University, Reading, UK. 384 pp.
Schwartz H J and Said A N. 1987. Sheep and goat nutrition on natural pastures in Kenya. In: Adeniji K O and Kategile J A (eds), Proceedings of the Workshop on the Improvement of Small Ruminants in Eastern and Southern Africa, Nairobi, Kenya, 18-22 August, 1986. OAU/IDRC (Organization of African Unity/International Development Research Centre), Nairobi, Kenya. pp. 111-125.
Semenye P P. 1987. Factors influencing Maasai cattle productivity and nutrition in Kajiado district, Kenya. PhD thesis, University of Nairobi, Nairobi, Kenya. 326 pp.
Wagenaar K T. Diallo A and Sayers A R. 1986. The productivity of transhumant Sudanese Fulani cattle in the Niger Inner Delta in Mali. ILCA Research Report 13. ILCA (International Livestock Centre for Africa), Add is Ababa, Ethiopia. 57 pp.
Wilson R T. 1982. Husbandry, nutrition and productivity of goats and sheep in tropical Africa. In: Gatenby R M and Trail J C M (eds), Small ruminant breed productivity in Africa. ILCA (International Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. 61-75.
