Previous Page Table of Contents Next Page

Breeding for milk in Kenya, with particular reference to the Sahiwal Stud

K. Meyn and J.V. Wilkins

Kenya's cattle population of over 9 million head is kept under extremely varying ecological and management conditions. The objectives of cattle husbandry are equally diverse. The ecological zones range from the Afro-alpine moorlands on Mount Kenya to the very arid areas in the north; they include the highly fertile humid and subhumid agricultural areas of the highlands, the semiarid and arid range areas, and the fertile coastal strip.

Farming systems range from subsistence agriculture and pastoralism to large-scale commercial farming and ranching. The objectives of cattle husbandry vary gradually from the traditional emphasis on numbers to higher production levels per animal and to the generation of income. The rapid increase of the human population and rising income expectations call for a better utilization of the cattle resource.

Several factors contribute to the low efficiency of Kenya's cattle herds: disease, overstocking, limited knowhow and capital, and a low genetic potential. The artificial insemination service in the areas with high potential and a regular supply of breeding stock for natural service in the range areas are efficient delivery systems for improved genetic material. To overcome the other constraints a much greater extension effort is required, which in turn will also emphasize the need to breed higher producing cattle.

Table 1. Ecology and cattle breeding policy for commercial and subsistence farmers in Kenya

Ecological zone1ClimateApprox annual rainfallPercent age of surface areaCattle populationStocking rate2Breeding policy
Subsistence smallholdersCommercial, large - scale
  Millimetres MillionsStock units/ha  
I to IIIAfro-alpine equatorial, humid to semiarid900213.72Dairying3(i)Dairying3
European breeds(ii) Dualpurpose European breeds4
IVSemiatid600–90092.53–4Subsistence beef5(i) Dairying5 Ranching6
European x zebu crosses(ii) Beef
VArid400–600501.94Subsistence: beef-SahiwalBeef-Boran
VIVery aridunder 400201.015Subsistence: beef-indigenous zebu-

1 From Pratt et al., 1966.
2 Stock units, equivalent to 450 kg liveweight.
3 Ayrshire, Friesian, Guernsey, Jersey.
4 Simmental, Brown Swiss, Red Poll.
5 Crisscrossing: Friesian, Simmental, Brown Swiss, Red Poll, with Sahiwal and Boran.
6 Commercial crosses: Charolais, Simmental, South Devon, Sussex, Hereford, Friesian, with Boran and Sahiwal.

K. Meyn is Livestock Specialist for the World Bank, Rockville, Maryland, United States: he was formerly Adviser on Cattle Breeding at the National Animal Husbandry Research Station, Naivasha, Kenya. J.V. Wilkins is Senior Livestock Officer at the Naivasha Research Station.

Ecology and genotype

In principle, high-yielding cattle can be kept anywhere in the world, provided enough capital and knowhow are available to create the necessary environmental conditions. The question that must be asked is whether this is feasible and economical. For the foreseable future, cattle owners in Kenya will have to think in terms of production systems in which cattle utilize the existing environment with relatively small capital inputs. Once a shortage of land begins, and when better technical knowledge has been accumulated, more intensive production systems can be considered.

The cattle breeding policy for each ecological zone and production system in Kenya has been formulated on the basis of research findings and some educated guesses (Table 1) about the requirements of the people and the potential of the environment to meet these, Permanent field recording is necessary to check whether the cattle breeding policy keeps in line with development trends.

I. Friesian - Sahiwal crossbred heifer at Naivasha shortly after calving, with two calves sired by a Simmental bull

High-potential areas

Over one third of Kenya's cattle are kept in the high-potential areas where small holders and large-scale farmers derive their main income from crops. Only productive cattle are competitive. The climate is suitable for European cattle, dairy breeds being favoured by the present milk/beef price ratio. The future outlook favours dual-purpose cattle, especially on large-scale farms. Rapid up-grading with Ayrshire. Friesian, Guernsey and Jersey from the indigenous zebu is taking place. However, results from a large wellmanaged farm in a high-potential coffee-growing area (1 100 mm average annual rainfall) indicate that Sahiwal/Jersey crosses outyield the Jersey (see Table 2). It is thought that purebred European cattle will outyield the Sahiwal crosses on a higher plane of nutrition, but surveys carried out on smallholdings in the high-potential areas indicate that the average yields per grade cow and year are in the order of 1 350 kg, i.e., well below the production level on this farm. Only a few top herds reach averages in excess of 4 000 kg milk per cow and year.

There are now about 800 000 socalled “grade” cattle in this zone which have more than 50 percent European blood. Less than 300 000 of these are kept on large-scale farms: the majority are kept by smallholders. Over 20 000 breeding females are currently being reached by the rapidly expanding Al service. Most of the AI bulls have to be imported. Currently, bulls originating from 11 countries and five continents are standing at the Central Artificial Insemination Station.

Despite the low production level, there is genetic variation. With improving management and feeding conditions this is expected to increase. For a number of reasons it has been found desirable to organize an AI breeding programme for the three leading dairy breeds, the Ayrshire, the Friesian and the Jersey. Production conditions are so different from those in developed countries that genotype-environment interactions cannot be ruled out. A progeny test trial to answer this question will commence soon. In the meantime, each dairy bull recruited by the Central Artificial Insemination Station undergoes a progeny test on government stations and parastatal farms. The aim is to get 30 heifer lactations for evaluation per test bull and to select one sire out of five tested for semen production. For the time being the test capacity of about 30 bulls per year and the rapid expansion of the AI service do not permit this selection intensity. Further expansion of the officially milk-recorded cow population from the present 7 000 head will be essential.

Table 2. Lactation milk yield and calving intervals of Jersey and Sahiwal Jersey crossbred cows on a mixed farm in a high-potential area of Kenya. 1969–72

 JerseySahiwal Jersey
Number of cowsMeanNumber of cowsMean
 Kg Kg
Milk yield per lactation    
First1531 0441881 446
Second1001 3711161 623
Third531 571541 771
Fourth221 780232 051
  Days Days
Calving internal    
First and second100359116371

Data collected by Wilkins, 1973

2. Brown Swiss-Sahiwal and Ayrshire-Sahiwal heifers (2 years) at Kilifi Plantations Ltd. in the coastal belt of Kenya.

For financial and genetic reasons it is desirable to recruit more bulls bred within the country for the AI service. Bull mothers are selected on performance and conformation, and are bred to imported semen from outstanding bulls of the breed. Bull calves born to contract matings are delivered to the Central Artificial Insemination Station at the age of four weeks after which they are reared under uniform conditions. Mild culling on growth performance is planned for later years. When the homebred bulls are ready for service, they enter the progeny test programme along with the imported bulls.

3. Lay-off bull with desirable beef conformation.

Semiarid and coastal areas

Most of these areas are occupied by subsistence farmers and pastoralists. Milk for home consumption and feeder steers for the market are the main products. Crops are too unreliable to provide a livelihood without livestock. Cattle crossbred between a European dual-purpose breed (Friesian, Simmental, Brown Swiss, Red Poll) and the Sahiwal are considered to make best use of the environment. While Meyn and Wilking (1973) reported that at the Ngong Livestock Improvement Centre in the semiarid Kenya highlands first lactation milk yields were 1 519, 2 152, 2 233 and 2 466 kg for Sahiwals, Ayrshires, Ayrshires/Sahiwals and Sahiwal/Ayrshires respectively, Wilkins (unpublished) found the production data on a well-managed farm in a semiarid area to be as set out in Table 3.

It has been concluded that a dairy herd of Friesian cows would produce slightly more milk than a herd of crossbreds, but the additional number of calves weaned would favour the crossbred unit in its overall economy. It was also reported that the Sahiwal bulls used in the herd were far inferior to those presently used at the AI station, which have been selected through the breeding programme. In a different study, Kimenye (1973) found that in a well-managed herd on the Kenya coast lactation milk yields and calving intervals for Ayrshire, 3/4-Ayrshire 1/4-Sahiwal, and 1/2-Sahiwal 1/2-Ayrshire cows averaged 3 024, 3 151 and 3 004 kg, and 424, 405 and 404 days, respectively, thus favouring the 3/4-bred Ayrshires. The formation of a new breed from a crossbred foundation is not considered, because of changing management and economic conditions. By the time a long-term selection programme yields its first results, farmers may need types of cattle different from those selected. By crisscrossing a European breed and the Sahiwal, farmers will benefit from genetic progress made in the selection programmes for existing breeds in the high- and low-potential areas or abroad. Flexibility in choosing the breed generation by generation will permit stockmen with a high standard of management and nutrition to increase the proportion of genes from the European breed, while under inferior environmental conditions farmers may decide to increase the proportion of Sahiwal blood.

Table 3. Mean performance of Friesians and Sahiwal/Friesians on a large-scale farm in Naivasha district, 1969–72

 Friesian FriesianSahiwal/Friesian
Lactation milk yield (kg)2 4952 341
Lactation length (days)322294
Calving interval (days)433410
Number of abortions per 100 adult females5.62.7
Calf mortality to weaning per 100 adult females8.53.5
Number of live calves weaned per 100 adult females74.889.0

Data collected by Wilkins, 1973.

Some large-scale operations are practising dairy ranching, i.e., milking their beef cows once a day. In the vicinity of a milk market this system is very economical. It is basically the same production system practised by the pastroalists, and the appropriate breeding policy would be crisscrossing a European dual-purpose breed with the Sahiwal.

Arid and very arid areas

The arid areas are of low agricultural potential, mainly occupied by seminomadic and nomadic pastoralists whose diet consists of milk, meat and blood. The 1.9 million cattle in these areas are kept for subsistence, milk and feeder cattle production. European cattle and their crosses with the Sahiwal probably have nutritional requirements that are too high for them to exist and produce in this environment. The Sahiwal is thought to be a suitable dual-purpose dairy/beef zebu for the low-potential areas. Breeding stock is disseminated for natural service by the National Sahiwal Stud at Naivasha, totalling about 80 bulls per year, and for artificial insemination by the Central Artificial Insemination Station at Kabete, using only proven sires.

4. A group of yearling bulls during their range performance test at Naivasha, Kenya.

The “Improved Boran” is bred only for beef production.

5. Calf-rearing unit at the National Sahiwal Stud in Naivasha — young calves are fed and sheltered at night in movable pens.

Environmental conditions are believed to be too harsh for any genetie improvement in respect of milk production among the 1 million indigenous zebus in the very arid areas. Natural selection for survival and fertility has probably produced the most efficient cattle population.

The National Sahiwal Stud

Zebu cattle with good dairy characteristics are desirable for purebreeding in the arid areas and for crossbreeding with European breeds in the semiarid areas. They may even be needed for crossbreeding in high-potential areas (see Table 2). Cattle for these purposes are being selected for milk and beef production at the National Sahiwal Stud at Naivasha.


Attempts to develop an improved indigenous dairy zebu in a number of livestock improvement centres came to an unsuccessful end in 1939 when four Sahiwal bulls were imported by the Government from Pusa, India, for an upgrading programme. By 1963, 60 bulls and 12 females had been brought from four herds in India and Pakistan. Early successes in upgrading the indigenous zebu (Mahadevan et al., 1962) were followed by disappointments. Despite careful pedigree selection, some bulls with an undesirable dairy temperament were brought in from all four places of origin. This resulted in a high percentage of their progeny failing to let down milk in the absence of their calves.

The formation of the National Sahiwakl Stud at Naivasha in 1962 concentrated the breeding work on one farm, and was the start of a scientific breeding programme for a dualpurpose zebu to be kept under extensive pasture conditions, but under good management. A progeny test programme for milk production commenced in 1965 following recommendations by Mason (1965). The current breeding programme has operated since 1968.

6. High-yielding bull mother of good dairy type, with desirable udder and teat shape.


The main objectives of the stud are:


The stud consists of 500 cows plus followers, a total of about 1 300 head. All heifer calves born and about 80 bull calves are reared per year. At birth the calves are immediately removed from their dams, weighted, and their identification number is tattooed in their ear. They are weighed weekly until they reach 55 kg liveweight, and thereafter fortnightly until 125 kg liveweight. They are dehorned with a hot iron at six weeks of age.

Calves are weaned from whole milk on to concentrates at nine weeks of age, having consumed about 210 kg milk in the rearing period. Concentrates are offered in the fifth week of life and they are consuming about 0.5 kg per day at the time of weaning. This gradually increases to 1.4 kg per day, and they are fed at this level until they reach 125 kg liveweight. Calves are reared on natural pasture. Only during times of drought do they receive lucerne hay as a supplement.

The calves are dosed with phenothiazine every time their pasture is changed and finally when they reach 125 kg liveweight. They are then branded on the left leg with the same identification number they received at birth. Bulls and heifers are sent to separate farms where they graze indigenous pasture without supplementary feeding for their remaining rearing period. They have access to a balanced mineral lick.

Bulls are selected at two years of age, either for progeny testing or for natural service in the range development areas. All heifers are inseminated at 27 months. Seasonal calving is not practised. Pregnant cows are drafted into the milking herd for one month and heifers for two months before calving. The longer period for the latter allows them to become fully accustomed to the milking surroundings. Small quantities of concentrates (0.5 kg) are on offer.

7. Breeding plan at the National Sahiwal Stud, Naivasha, Kenya.

All cows and heifers are weighed immediately after calving. They are bred again 70 days after parturition and are dried off at 305 days of lactation. The herd is milked in the field twice a day through four mobile bails, each manned by six hand milkers and a recorder. Individual yields are recorded at every milking and butterfat analyses are carried out once a month. The indigenous pasture is expected to provide for maintenance and the production of 5 litres of milk per day throughout the year. After prolonged rain, concentrates are fed only to cows exceeding 9 kg milk per day, at the rate of 1 kg for 2.5 kg. milk produced.

The milking herd has the best pasture available. Good paddocking permits rotational grazing, with the cows rarely staying longer than seven days in one paddock. Dry herds follow the milking herd to eat the less palatable grasses.

Productivity and genetic parameters

The collection of production data over a number of years has permitted the estimation of genetic parameters. Table 4 gives the averages, standard deviations and heritability of some production traits.

The existing variance and the corresponding heritability estimates give promise of a rapid genetic improvement in respect of milk yields, and to a lesser extent in respect of growth rates. This is also demonstrated by the increase in yields since the systematic selection programme commenced in 1968 (see Table 5).

Breeding plan

The breeding plan for milk and beef is given in Figure 7. About 210 heifers enter the milking herd every year. The higher yielding 110 heifers are allowed a second lactation, after which the best 55 cows over the first two lactations are retained in the herd. Thereafter, hardly any culling is done on yield.

Table 4. Least square means, standard deviations and heritability estimates for some production traits of Sahiwal cattle at Naivasha, Kenya

TraitNumberLeast square meansStandard deviationsHeritability estimates
Birth weight (kg)    
1 51521.42.50.14
Average daily gain to 125 kg liveweight (g)    
1 424526700.07
Bulling heifer weight at 27 months (kg)944307370.30
First calving age (months)85137.34.5 
First lactation    
Length (days)
Yield (kg)
Butterfat (%)
Butterfat (kg)

Source: Meyn and Wilkins (unpublished).
1 From 1963 to 1965 birth weights have been recorded only for bull calves scheduled to be reared for breeding
2 Butterfat samples take only from cows yielding over 3 kg milk per day.
3 The corresponding heritability estimate for milk yield on the reduced material was 0.19.

Table 5: Lactation lengths and milk yields at the National Sahiwal Stud, Naivasha, 1965–72

YearFirst lactationsSecond lactationsThird and later lactations
Number of cowsLengthMilk yieldNumber of cowsLengthMilk yieldNumber of cowsLengthMilk yield
  DaysKg DaysKg DaysKg
19651491847101492291 1021982391 246
19661362591 0011282611 1312822761 382
19671212451 0421292471 0713352631 280
19682332509591062601 1753532661 317
19691822551 011882861 5341762901 763
19701712611 057742961 7701542971 934
19712242451 143702991 7691802921 910
19721442771 3061252711 6891872861 840

Source: Wilkins et al. (unpublished).

Heifers and first calvers are inseminated with semen from test bulls, while all cows calving for the second time or more — the so-called elite herd — are inseminated with semen of proven bulls to produce the next bull generation. The 180 elite cows produce 150 calves, i.e., 75 bull calves per year. Of these, about 70 survive the bucket-rearing period and range-performance test for growth rate to an age of two years. The best 15 are then identified by a selection index, comprising the estimated breeding values for milk yields of the sire and dam, and the estimated breeding value for weight for age of the young bull itself. An inspection of the selected bulls and their dams for physical faults, muscling, udder conformation and teat shape eliminates three more bulls. The remaining 12 are tested for semen quality, leaving about 10 for test mating. The aim is to produce 14 milking daughters per test bull.

The average daily gain to weaning, the bulling heifer weight at 27 months, the 30-day milk yield and the 305-day milk yield of the first lactation are analysed by contemporary comparison. First decisions on test bulls are made when the 30-day yields become available. Bulls with outstanding daughter performances are sent to the AI station for semen production, while all the remaining bulls are kept on. Final decisions are made on the results of the 305-day milk yields. Two bulls are selected annually for the AI station and for use in the elite herd. The remainder are sold for natural service or for slaughter. Genetic progress in terms of milk yield may be expected at 3 to 4 percent per year (Meyn, Were and Bartilol, unpublished).

In recent years the demand for Sahiwals from the range areas of Kenya and from abroad has risen sharply. Deep-frozen semen or breeding stock have been exported to many countries, mainly in tropical Africa. There are plans to extend the Sahiwal breeding programme to other government farms. This could lead to even faster genetic progress.


The national dairy cattle breeding policy for the different ecological zones of Kenya is based on findings which suggest that, with adequate management, best use of pasture is made by European dairy breeds in the temperate high-potential areas, by crosses between European breeds and the Sahiwal in the semiarid and coastal areas, by pure Sahiwals in the arid areas, and by indigenous zebus in the very arid areas. The Sahiwal has a very important role to play, both as a purebred and as a crossbred with European breeds.

With the existing variation and the corresponding genetic parameters, rapid progress may be expected in the breeding programme of the National Sahiwal Stud at Naivasha. The overall herd average has increased from 1 042 kg milk per cow and lactation in 1965 to 1 630 kg in 1972. A future rate of genetic gain of 3 to 4 percent per year has been projected. Sahiwal breeding stock and deepfrozen semen are being exported to many tropical countries.


Kimenye, M.D. 1973. Comparison of Ayrshire and Sahiwal crossbred cows with high grade Ayshires on Kilifi Farm, Coast province of Kenya. M.Sc. thesis, University of Nairobi, 95 p.

Mahadevan, P., Galukande, E.B. & Black, J.G. 1962. A genetic study of the Sahiwal grading-up scheme in Kenya. Anim. Prod., 4: 337–342.

Mason, I.L. 1965. Report to the Government of Kenya on the National Sahiwal Stud. FAO Report No. 1965. 27 p. Rome, FAO.

Meyen, K. & Wilkins, J. V. 1973. National cattle breeding programmes in Kenya. Paper presented at the third World Conference on Animal Production, Melbourne 1973, 5(e)-1–10.

Meyn, K., Were, H. R. & Bartilol, P. Studies on Sahiwal cattle in Kenya. 2. The breeding programme for the National Sahiwal Stud. (To be published)

Meyn, K. & Wilkins, J.V. Studies on Sahiwal cattle in Kenya. 4. Genetic and environmental influences on preweaning weights and gains of calves. (To be published).

Meyn, K. & Wilkins, J.V. Studies on Sahiwal cattle in Kenya. 5. Genetic and environmental influences on first lactation performance. (To be published)

Wilkins, J.V., Siele A.K., & Shililu, L. Studies on Sahiwal cattle in Kenya. l. The National Sahiwal Stud, its history, functions and management. (To be published).

Pratt, D.J., Greenway, P.J. & Gwynne, M.D. 1966. A classification of East African rangeland. J. appl. Ecol., 3: 369– 382.

Previous Page Top of Page Next Page