Introduction

The German East Africa colony was among the earliest African countries to attempt improvement of the production of native cattle. These activities, begun in 1905, were mainly directed to the control of epidemic and endemic diseases including Foot and Mouth Disease, Contagious Bovine Pleuro-Pneumonia, East Coast Fever and Trypanosomosis (Crabtree 1917; LPRI 1983). Cross breeding of exotic animals on the indigenous Tanganyika Shorthorn Zebu (TSZ) (Bos indicus) became an important strategy after the former colony became the Tanganyika Territory in 1918, as a League of Nations Trusteeship under British tutelage. As early as 1926, a variety of Zebu bulls (Bos indicus) was imported from India, as well as Bos taurus types including the Afrikander from South Africa and Angus, North Devon, Friesian and Ayrshire from the UK (DVSAH 1926). The intention was to improve the TSZ — but more particularly the Iringa Red variety (Fig. 1), considered to be the “North Devons of Tanganyika Territory” – in respect of both milk and meat, as the local people had “paid more attention to the size of the humps of these indifferent milkers than to the shape of their rumps” (DVSAH 1926).

Fig. 1
figure 1

Native Iringa Red cow with halfbred Afrikander calf at the Tanganyika Territory Government Stock Farm at Puku, Dar es Salaam in 1926 (Source: re-photographed by the Author from DVSAH 1926)

Full size image

This paper is an attempt to review the often confusing and sometimes confused origins, development, production and distribution to farmers of the so-called breed of Mpwapwa cattle.

Methodology

This paper is based in part on the author’s association with Tanzania where he worked full time from 1961 to 1970 and has since made numerous consultancy visits up to 2013. During this 52-year period, he was intimately involved with livestock production and development with research scientists and with smallholder farmers and pastoralists.

The paper derives largely, however, from a detailed analysis of the literature relating to Mpwapwa cattle. Data bases were searched using the key words “Mpwapwa” and “Tanzania”. Such a search failed to identify other papers relating to the cattle that did not have these key words but many items have been incorporated based on reference lists in published papers and on the author’s own knowledge and experience. The result is a comprehensive but certainly not complete bibliography of Mpwapwa cattle.

Development of Mpwapwa cattle

Following the UK taking up the Trusteeship of Tanganyika Territory, local TSZ cattle were introduced to the newly established Mpwapwa Livestock Station (the headquarters of the nascent Veterinary Department) in central Tanzania in the early 1920s for production of anti-rinderpest vaccine (DVSAH 1927). Early attempts at making the TSZ more productive were made with the import, initially to Puku near Dar es Salaam, of a range of bulls of both Bos taurus and Bos indicus from several sources (Fig. 2).

Fig. 2
figure 2

Bulls imported from various countries at Puku in 1926 to be used in improvement of native TSZ cattle (Source: re-photographed by the Author from DVSAH 1926)

Full size image

What should have been a structured breeding programme began in 1935 (DVSAH 1937) and has continued intermittently and with many changes of policy and practice to the present day (Fig. 3). At that time, B. taurus bulls, mainly UK Ayrshires, were used on local cows with the aim of producing a new breed adapted to the local environment but of higher output than the native stock (French 1940; MWLD 2004). Some five years into the programme, it was realized that crossing to B. taurus was failing to create the hoped-for animal. The breeding plan was thus modified to introduce B. indicus (Red Sindhi and Sahiwal) genetics, imported from South Asia (now India and Pakistan) to replace at least partly the B. taurus genes. Frequent changes in research personnel were invariably accompanied by changes in the breeding scheme which, it has been said, was “not well documented” (Kiwuwa and Kyomo 1971). One example of change was the cancellation of the programme in 1940 as the results did not come up to expectations (French 1940, 1941). One source that ought to be authoritative, ignores the early experiments and states that the Mpwapwa “was developed from the original concept of an Indo-African breed by Dr H.G. Hutchison who first crossed the local Tanganyika Zebu cows with Red Sindhi and Sahiwal bulls imported from Kenya and called the cross ‘Indo-African’” (AGTR 2020).

Fig. 3
figure 3

A timeline of the development of Mpwapwa cattle (Source: constructed by the Author)

Full size image

A new programme was started in 1944. It was now the intention to retain 30% of the genetics of the 1940 type (i.e. B. indicus Indo-African Zebu with traces of B. taurus inheritance) in the final composite, together with 60% of Sahiwal and Red Sindhi and 10% of Ayrshire (Syrstad 1990). The meandering path traced over the years reflects, all too often, the interests of the current researcher rather than the long-term development of a truly fixed genetic type. Thus, following the initial period, the blood of Jersey, Guernsey and Friesian was introduced as was that of Boran, Ankole and Sahiwal. The herd at Mpwapwa station was closed in 1956 to enable selection to be carried out (Getz 1974). In 1958, the cattle were said to comprise 35% of Red Sindhi, 20% Sahiwal, 20% TSZ, 10% Boran, 5% Ankole and 10% B. taurus “mainly Ayrshire” (Macfarlane 1971; Rushalaza and Kasonta 1993; Kurwijila 2001). At the same time, this less than distinctive animal was accorded breed status as the Mpwapwa.

A breed improvement programme was instituted at the same time as the promulgation of breed status as was a progeny testing scheme. The intention was to produce a dual purpose animal for the semiarid environment of central Tanzania with a withers height of 119 cm, an adult male weighing 520 kg and a female 400 kg with a calving interval of 380 days, capable of producing 2300 kg of milk in 305 days and a steer carcass of 230 kg in less than four years (Macfarlane 1971). A research officer who worked at Mpwapwa in 1961–1964, in a personal communication to the author of this paper wrote, on 30 December 2020, “in my time at Mpwapwa if there was a written, constructive breeding project and defined breeding programme I never saw it”. A combination of adverse circumstances in 1965, including Johne’s disease eradication procedures and a senior staff shortage resulted in a return to uncontrolled multiplication as it had been prior to 1958 and brought breed development to a halt (Katyega 1987; Getz et al. 1986).

Over the succeeding half century, there have been multiple further changes to the breeding plan. In 1963, the animal was still very variable in physical type (Fig. 4) and in production traits. In 1968, 10 years after the declaration of breed status, individual animals varied in genetic background carrying from 3 to 88% Red Sindhi inheritance, 0–69% Sahiwal, 0–63% TSZ, 0–59% Boran and 0–34% Ayrshire (Mchau 1988; Syrstad 1990). From 1968 to 1971 some Mpwapwa females, of clearly mixed B. indicus and B. taurus blood were mated to Friesian, Ayrshire, Guernsey and Jersey bulls to produce a crossline, the females of which were then backcrossed to Mpwapwa bulls (Mkonyi 1982). Breed development continued again and in 1971, the herd at Mpwapwa was said to be 32% Red Sindhi, 30% Sahiwal, 19% TSZ, 10% Boran and 9% Ayrshire and Shorthorn (Kiwuwa and Kyomo 1971; Katyega 1987). It was decided in 1974 to further develop the animals into distinct milk and beef lines (LPRI 1976, 1977). This initiative was curtailed in 1977, however, following a directive from the veterinary headquarters (Getz et al. 1986).

Fig. 4
figure 4

Mpwapwa cattle at Mpwapwa Research Station central Tanzania five years after declaration of breed status in 1958 (Photograph by the Author, November 1963)

Full size image

In 1984, the Director General of the Tanzania Livestock Research Organization quoting the National Dairy Development Plan stated that Tanzania aimed to “increase dairy production at the most rapid rate possible, with the least cost and on a long-term basis. Development is aimed at eventual self-sufficiency in dairy production and a consumption pattern which ensures that minimum nutrition levels are attained by all sectors of the community” (Macha 1984). As a part of the Dairy Development Plan, breeds for general use would be restricted to Jersey, Friesian, Ayrshire, Sahiwal and Mpwapwa and in areas with greater problems of heat stress or high disease challenge the zebu dairy breeds of Sahiwal and Mpwapwa and crosses of these would be the predominant breeds (Macha 1984). At the same time, the Dean of the Faculty of Agriculture, Forestry and Veterinary Science of the University of Dar es Salaam (now Sokoine University of Agriculture) wrote, “studies have shown that grading up to Sahiwal leads to further increases in milk yield” and “In Tanzania, the use of a three breed cross - Red Sindhi, and Sahiwal cattle (both from India) and the local zebu -- as well as the use of other genes such as Boran and Ayrshire have resulted in the production of a dual purpose animal, the ‘Mpwapwa’, for beef and milk” and that “These have been extensively used by dairy farmers in medium altitude and coastal belt areas of Tanzania” (Kyomo 1984).

Natural mating has been used during most of the long period of breed development. General practice after 1958 was a single sire herd with some hand mating. Random multi-sire herds with selected young sires were introduced later in an attempt to decrease the generation time. In these later herds, cows were supposedly of balanced strata of age and lactation status. Considerable use of Artificial Insemination was made, however, at the end of the 1960s and in the early 1970s. During this period many “pure” Mpwapwa cows were mated to European dairy sires but this practice and inaccurate recording of sire breed resulted in the loss of at least one generation in the progression to a stabilized and relatively reliable breeding population (Getz et al. 1986). The breed was still being developed in 1990 when an open nucleus scheme was attempted (Kasonta 1989; Kasonta and Nitter 1990) and when it was identified as “an Indo-Euro-African synthesis” (Syrstad 1990).

In 2001, “Reliable information indicates that currently there are less than 100 breeding females in the various research stations in the country. Further conservation work for the breed has been in the pipeline for sometime now, but so far no support has been forthcoming” (Kurwijila 2001). There were probably fewer than 500 breeding cows in existence in 2002 when a multiplication programme was mooted with assistance from the Tanzania/Japan Food Aid Counterpart fund (Kurwijila and Boki 2003). At this period, the Mpwapwa was described as “an established breed of four Bos indicus breeds and 8 per cent of Bos taurus” and as a dual-purpose dairy breed (Bwire and Wiktorsson 2003). Tanzania requested purchases of Sahiwal bulls from Kenya on at least 10 occasions between 1971 and 2007 (Mailu et al. 2012). In 2004, in its Country Report to FAO, Tanzania stated that the Mpwpwa breed was “at risk of being lost” although there was an in situ conservation programme operating at Mpwapwa station (MWLD 2004). Some years later, in 2010, there were said to be between 1500 and 3000 Mpwapwa cattle including ten breeding bulls and 1000 cows with 95% of females being bred pure, that numbers were increasing, that there was 160 herds with an average number of animals per herd of 12, that Artificial Insemination (AI) was used with semen from four bulls and that the breed was not at risk (FAO 2020). The “risk of being lost” status was reiterated by the Director of Livestock Production in 2013 (Njombe 2013). Bulls are kept at the National Artificial Insemination Centre near Arusha (Msalya et al. 2017; Porter 2020). There is very little uptake of semen by smallholder farmers and pastoralists and medium and large scale commercial dairy farms prefer to use recognized dairy breeds, with the Ayrshire still being a preferred type (Njombe 2013).

There has been very little assessment of any genetic improvement. What has been done has been confined to the cattle at Mpwapwa station. It has, however, provided sufficient material for two postgraduate studies (Mpiri 1982; Kasonta 1988). The most comprehensive study used data from 1967 to 2012 including 1003 lactations from 385 cows and 78 sires. Heritability estimates for lactation yield (1480.0 ± 506.3 kg) and 305-day yield (1686 ± 578.6) were moderately high but low for lactation length (271.4 ± 44.7 days). Repeatabilities for yield and 305-day yield were high but moderate for lactation length. Heritabilities for age at first calving (32.0 ± 7.1 months) and calving interval (493.1 ± 130.7 days) were low as was repeatability for calving interval. Variation among estimated breeding values was significant and led to a suggestion that selection to improve these traits was feasible and selection for specific sires could result in increased lactation yield. Periods of years within the data showed some increments and some decrements but in the long run all traits showed a decline in genetic progress in the on-station breeding programme (Chawala et al. 2017a, 2017b).

Production parameters

On-station research

On the Mpwapwa station, the milk yield according to one source was about 1600 kg (Das et al. 1986). Yield varied, when short lactations were excluded (sic), from 1447 to 1628 kg with an age at first calving of 45.4 months and a calving interval of 447 days according to another (Syrstad 1990). Lactation curves and persistence were similar to most types of Zebu (Mchau 1988; Mchau and Syrstad 1991). In one trial, birth weights of Mpwapwa cattle with 12% B. taurus inheritance were lighter than backcrosses with 34% B. taurus blood and were lighter at 252 days, but there was no difference in weights at older ages up to 752 days. There was no significant interaction between the Mpwapwa and the backcrosses born in the wet or dry seasons. It was concluded that the backcrosses were not more sensitive to the nutritional stress of the dry season than were the pure Mpwapwa calves (Msanga et al. 1986).

In another study, 20 purebred Mpwapwa (said to be 92% B. indicus — 32% Red Sindhi, 30% Sahiwal, 19% TSZ, 11% Boran — and 8% B. taurus mainly Ayrshire) cows and 16 crossbred Mpwapwa cows (derived from 66% Mpwapwa and 34% Ayrshire, Friesian or Jersey B. taurus breeds (Macha 1986) and their young were used in the experiment which compared restricted suckling and artificial rearing of calves. Production of saleable milk was increased by 15.0% in the Mpwapwa and by 34.3% in the crossbred cows under restricted suckling. Restricted suckling considerably extended the length of postpartum anoestrous and calving interval in both types of cow, and the incidence of mastitis was reduced to minimum levels compared to full suckling. Calf daily gain was similar for restricted suckling and artificial rearing of calves. It was concluded that by applying restricted suckling, the availability of milk for human consumption was increased by reducing the competition between people and calves (Mejia et al. 1998).

Several small studies on nutrition and nutritional effects on Mpwapwa and crossbred cattle have been made (Bwire and Wiktorsson 1996, 2003; Das et al. 1999). There was also sufficient material for one postgraduate dissertation (Bwire 1995). Studies comparing the performance of Mpwapwa cattle and domestic buffalo have been undertaken at Mabuki Farm south of Mwanza in western Tanzania (Katyega et al. 1980; Katyega 1981).

On-farm studies

Between 1971 and 1985, 270 bulls and 86 females were sold to other parastatal organizations or to private farmers and villages (Syrstad 1990). It was not, however, until 1986 that on-farm (as opposed to the previously unique on-station) work started in attempts to broaden the data available for further development (Kasonta and Mkonyi 1991; Rushalaza et al. 1993). By the late 1990s, however, there were no “purebred” animals on farms outside Mpwapwa station and the breed was once again on the verge of extinction (Rege 1999, 2000; Rege and Tawah 1999). One more recent report states that from January 2006 to December 2011, some 426 bulls and 137 heifers were distributed from Mpwapwa Livestock Research Centre to 72 breeders’ associations and nine individuals (TDN 2012).

An on-farm technology package introduced at Berege village included improved Mpwapwa cattle, improved forage species, intensive feed gardens, conservation and use of crop residues, disease control and dual purpose legumes. Collaborating farmers purchased the improved cattle and adopted the other technologies at varying levels with none of the latter being rejected completely. Based on a questionnaire survey it was said that the introduced cattle performed as well as on-station. Averages of production coefficients were 1626 kg milk per lactation of 240 days, 498 days calving internal and 33 months age at first calving. It was concluded that there was an urgent need to introduce an appropriate breeding programme for the village in order to avoid the increased possibility of inbreeding and to facilitate a wider distribution of the improved zebu type of cattle to farmers in who were in need of them (Kasonta et al. 1990; Rushalaza et al. 1993).

In one small study undertaken in two villages within 40 km of Mpwapwa and by use of a questionnaire, the milk production of 12 Mpwapwa and 12 crossbred cows was better than that of 45 TSZ cows, the crossbreds producing 4 l of milk a day when supplemented on natural grazing and 6 l when zero-grazed. The strong conclusion (no statistical analysis was done) was that, improved Mpwapwa cattle and their crosses perform better than local cattle and that more effort was needed in their multiplication and distributions. It was also concluded that better extension services were required which must go hand in hand with improvement of pastures and other feed resources together with improved disease management and market information through a concerted effort by research and extension (Komwihangilo et al. 2009).

A small number of Mpwapwa cattle have been distributed to smallholders in southern Tanzania (Komwihangilo et al. 2010, 2013).

Discussion

After almost 90 years of indecision with regard to support for and the status of research into livestock development and the on and off formal and informal research, the Tanzania Livestock Research Institute (TALIRI) was established as a body corporate by Act No. 4 of 2012 published in the Government Gazette of 27 July 2012. The Act provides for the functions and powers of the Institute in relation to the conduct of livestock research. Since its establishment, the Institute has been re-organizing itself through strengthening administrative and governance infrastructure “despite carrying [out] its core functions as stipulated in the Act” (TALIRI 2020). It would appear that with regard to research on Mpwapwa cattle, there has been less interest in the recent past than in earlier periods, emphasis having shifted to a considerable extent to direct improvement of the TSZ and of sheep and dairy goats. Only seven of 56 “previous projects”, four of 28 “ongoing projects” and none of 11 “new projects” relate to Mpwapwa cattle (Appendix), and there have been very few formal, peer-reviewed, publications by research staff in the period 1993–2017.

The aimed-for production parameters for dairy traits have never been achieved although steer weights at 4 years have been as hoped for. The less than targeted performance over the long development period is due to many factors. The breed multiplication plan has not been efficient owing to, among many other problems, a high turnover of animal breeding scientists and disease outbreaks including an abortion storm from 2005 to 2010 (Shirima et al. 2014). Other problems included fluctuating feed supplies (both quantity and quality) and numerous policy changes (Wilson 2009). Breeding activities have not been constant over time and at various periods have received emphases on dual purpose production, milk production or beef production goals (Syrstad 1990; Bwire et al. 2005). The original target of 2300 kg milk per lactation of 305 days has never been reached, nor indeed hardly approached. Declining milk production between 1985 and 1993 can be attributed to limited funding for research and production services due to the structural adjustment policies imposed by the International Monetary Fund and the World Bank to ensure debt repayment and economic restructuring. In common with many other developing nations, Tanzania reduced its spending on health, education, research and development activities. Most parastatal farms were reduced in size or privatized (Heidhues and Obare 2011). The resultant smaller number of breeding cows was followed by reduced selection pressure on both cows and bulls. Consequent on the failure of the public expenditure cut policies of the IMF/WB, the government and other donors resuscitated earlier efforts to expand parastatal activities. An upward trend in milk yield from 1994 can thus be explained by active government support for breed multiplication through projects such as the Tanzania Agricultural Research Project Phase II (TARPII) (World Bank 2004) and the Agricultural Sector Development Programme (URT 2006).

It is clear, however, that milk yields overall did not increase over a period of 45 years (Chawala et al. 2017a, b). It has always been difficult to maintain a number of animals sufficient for genetic progress to be made. The analysis of only 1013 lactations over 45 years from 1967 to 2012 means, for example that only 22.5 lactations per year were available; similarly there were only 10.3 calving intervals per year and 11.7 ages at first calving (Chawala et al. 2017a, b). Management strategies need to be revised especially in terms of general husbandry and feeding strategies. The genetic improvement programme also needs to adopt new technologies of feeding, reproduction and genotyping to improve the environment, fertility and selection of cows and bulls with an acceptable range of breed composition.

Composites have several advantages over rotational crossbreeding. It is a simple system that can still enjoy most of the heterosis benefits on fertility and longevity. Composite cattle can be created to fit many situations including market traits, adaptation, conformation, colour and fertility. The genetic merit of composites relies, however, on performance-oriented breeders who are committed to carefully planning the development of the composite line. The development of Mpwapwa cattle has singularly failed in this respect.

Conclusions

The number of Mpwapwa cattle has never exceeded 1000. Most of this mishmash of genetics has been kept on the station of its conception and, to a much lesser extent, on one other government station. Overall, very few individuals have “escaped” into the real world. Local farmers have been largely uninterested in making use of the type because of its minimal better performance over the TSZ and the extra costs incurred in labour, feed and veterinary care. In spite of a clear lack of success, public service researchers (and international organizations) have continued to spend good money on “development” while concurrently lamenting imminent demise. Although the breed has been described as threatened with extinction, at least since 1990, no coherent strategy for its dissemination seems to have emerged (Njombe 2013). Early in the twenty-first century, the “breed” may have contained as much as 75% Sahiwal blood (FAO 2020). In practice, the population does not have the identifiable and stable phenotypic characteristics (Fig. 5) nor the regular production traits that are generally required for “breed” status. In fact, there does not appear to be a documented “breed standard” precising measurable morphological and productive criteria, without which breed selection work cannot be carried out.

Fig. 5
figure 5

Part of the herd at Mpwpwa station in the early twenty first century (Source: front cover photograph of the “Agricultural Sector Development Programme Phase II (ASDP II)”. 2017. Ministry of Agriculture, Dar es Salaam

Full size image

The credentials of the Mpwapwa type as a breed have been enhanced by its being recorded by FAO as ‘Endangered’ (Scherf 2000). In a major 2007 publication on the state of the world's animal genetic resources by FAO, however, the Mpwapwa does not get a single mention (Rischkowsky and Pilling 2007). The Research Officer cited in the foregoing also wrote that “after 90 years, there is little to show for all the money and effort, and the Mpwapwa is an indiscriminate mix of breeds but Sahiwal being the main contributor”. One eminent livestock scientist and specialist in animal genetic resources, in a personal communication to me in January 2020, believed, however, “that even if it was derived from distinct breeds the Mpwapwa had probably existed as a recognised, selected and distinct entity for long enough to be acceptable as a breed under current definitions”. The author of this paper begs to differ: it is highly unlikely that the current exact genetic composition of the group is known or will remain stable. In addition, without a defined and accepted “breed standard” the Mpwapwa has never, ever, been a “breed”.