Previous Page Table of Contents Next Page

Practical technologies and options for the genetic improvement of livestock in developing countries

by L. Vaccaro & D.E. Steane


This paper considers some of the implications for breeding strategies regarding the need to promote sustainable animal production systems in the developing world. Due to restrictions of space, cattle in the lowland tropics are taken as the main example and it is hoped that many of the same considerations will be applicable to other species. The choice seems justified on the grounds of the numerical importance of cattle in the tropics and the fact that so much more needs to be known about tropical lowland, compared with temperate, zone systems in all fields including animal breeding. Since such large areas of the lowland tropics are characterized by acute poverty and hunger that the case also well illustrates the need faced by developing countries to ensure immediate increases in cheaply priced food whilst, at the same time, conserving their natural resources.


Sustainable cattle production systems in the lowland tropics must contribute to the alleviation of poverty, hunger and national indebtedness in both the short and long term. Priority must be given to those systems which offer the best opportunities for providing cheap food and other products. This means that they will be based primarily on locally abundant feed, animal and human resources. The improvements proposed must also be widely applicable and not directed solely to an economic or technical elite, otherwise they will not contribute fully to rural development.

These restrictions immediately define certain biological aspects of the systems which should be given priority, as well as also important guidelines for genetic improvement. Of the biological aspects, the nutritive basis of the systems is perhaps the most important single feature because it determines the type of animal which must be used. To comply with the objectives outlined above, the base feeds involved will be those which are locally abundant. Throughout the tropics, these will include agro-industrial by-products, crop residues and, in some regions, tropical grasslands. Grains, particularly imported cereals, will be restricted and, in general, concentrate supplementation reduced to the strategic use of specific nutrients to optimise ruminal function and the efficiency of use of the diet as a whole (Preston and Leng, 1987). The use of by-products and grasslands through animals contributes to the sustainability of the whole production system. Furthermore, animals can use wastes which have little alternative use and simultaneously provide traction and fertiliser, if needed. They also contribute to sustainability in the direct economic sense by providing a source of low-risk savings and, if milked, of daily income (Winrock, 1978). Tropical Latin America alone has over 500 million hectares of grasslands, often occupying marginal acid soils or steep slopes (Seré and Jarvis, 1989). Grazing systems under these conditions have a key role to play in soil erosion control and fertility improvement, particularly when legumes are included. In addition, the improvement of grazing systems already established on cleared tropical forest should help to stem the rate of forest destruction by providing a better livelihood from the land which is already cleared.

These considerations have important implications from the genetic point of view. In the first place, whether by-products or tropical grasslands provide the basis of the diet, their nutritive value is low and the levels of individual production will not be high. Expressed in terms of milk yield, perhaps 4 to 10 kg/cow/day are reasonable limits. Yields much above this level will involve the use of resources (e.g. imported grains, heavy capital investments, advanced technical skill) which prevent the systems from meeting the socio-economic objectives originally set out. Thus, the genetic potential of the cattle for production must be carefully matched to the resources available and, in this respect, the temperate zone criteria are irrelevant. Secondly, all improvement measures, including genetic ones, must be simple and cheap enough to be widely applicable and the selection of appropriate measures must ensure that they make the best possible use of the resources involved. These same principles appear to be equally relevant to other animal species, whether ruminant or non-ruminant. One of the most unsatisfactory features of pig and poultry production systems in many parts of their tropics is their total dependence on imported stock, technology and, often, feeds. This disregards the potential of small animals as contributors to rural development by exploiting their ability to make use of local feedstuffs, including wastes, and frequently represents an incomplete and unsustainable production system of which the feed base (e.g. imported cereals) is highly dependent on fossil fuels.


One possible strategy for genetic improvement is breed substitution. Defined as the introduction of commercial females, as well as males, this seems to be the least satisfactory option, if the strict socio-economic objectives are borne in mind.

In the first place, if the constraints of the local environment are correctly assessed and the products required from the animals precisely determined, it may well be that existing genotypes are adequate and able to respond sufficiently and economically to improvements in the system. Frequently, the complexity of the required product is often overlooked. Pigs, for example, play an important role in many small farming systems as a low-cost and, therefore, low-risk source of savings. Once exotic genotypes are introduced, purchased inputs become essential and risks increase to a level at which the original objective is lost. In the case of cattle, many lowland tropical systems require them to produce fuel, fertiliser and traction, besides meat and milk. Even where the first three are less important, as in most of Latin America, there are still strong arguments in favour of dual purpose, as opposed to specialized meat or milk production (Preston, 1976; Seré and Vaccaro, 1985). These multiple functions of cattle are an intrinsic element of the stability of the whole production system, which may well be upset if the existing stock are replaced. Resistance to disease and parasites may also be of vital importance. The trypanotolerance of West African N'Dama cattle is a case in point and increasing attention may be expected to be given in future to parasite resistance due to the high cost of insecticides and environmental concern. Tick resistance which is a characteristic of zebus or resistance to cattle flies such as Dermatobia hominis found in Latin American criollos (such as the Colombian Blanco Okejinegro) will also take on increasing economic importance.

Secondly, even where existing breeds have insufficient production potential, their substitution with new genotypes is, at least in the case of large species, a high-cost, high-risk measure which cannot be expected to solve the problems of the majority of farmers in a given locality. With cattle, neither substitution with another tropical genotype nor a temperate breed seems likely to be justifiable. Few tropical breeds, or crosses, have sufficient numbers of genetically evaluated stock to justify importation, even presuming that their performance in the new environment could be predicted. Whatever their success, the possibility seems remote that the introduced animals will have any advantage over the crossbreds which could be derived from the existing population, except in terms of the time required to produce the crosses; which is, however, relative. Usually overall productivity of the existing animals is limited by poor reproduction and survival. Investment in measures to improve these traits will have an important permanent impact on the efficiency of the production system and will pave the way for the higher potential crossbreds to be bred while the improvements are being organized. It would seem, therefore, that investment in environmental improvements along with crossbreeding is more likely to prove a cost effective alternative than the introduction of new tropical stock.

The introduction of European breeds into the lowland tropics is also unlikely to be a suitable option. Tropical forages are less digestible than temperate ones (Minson, 1980) and their utilization therefore generates more metabolic heat. As European cattle do not have efficient heat dissipation mechanisms, high levels of concentrates are required if they are to survive and produce. The concentrates are typically derived from cereals, often imported. These systems are clearly less sustainable in the long term than grazing systems or those based on by-products, for which pure European breeds are generally unsuitable. Furthermore, past evidence suggests that European dairy cattle will be unable to maintain their herd numbers, due to short and involuntary culling (Vaccaro, 1990). No genetic programme can be sustainable if it relies constantly on the importation of females. Besides, the costs involved in terms of concentrate feeds, veterinary supplies, technical skill and capital investment remove systems based on European cattle from the grasp of the ordinary farmer and prevent them from making the desired impact on rural development. Recent Latin American experience also questions their potential for providing low-cost food, since milk production costs have been found to be higher and profitability lower than with crossbred stock (Table 1). It is extremely likely that the same arguments will apply to the introduction of exotic breeds of small ruminants and also, to some extent, to pigs.


A second option to be considered is crossbreeding. In the case of cattle, evidence from all over the lowland tropical world shows that, in one generation, tropical populations can increase yields to levels very close to the limits set by the locally available feed resources and which fit conveniently with the levels of technical skill, capital investment and service support infrastructure available.

Despite this, there are few examples of successful, stabilised crossbreeding schemes in commercial populations on a large scale, and many cases of patent failure - usually involving grades of European crosses above the levels which the local environment can sustain. This situation cannot be attributed to any lack of discussion of the theoretical merits of different crossbreeding schemes for tropical cattle, at least in the context of milk production (Cunningham and Syrstad, 1987; Bondoc et al., 1989). The difficulty appears to lie in the fact that the field success of crossbreeding schemes depends far more on their practical feasibility than on their theoretical merits. It is suggested therefore, that genuinely sustainable crossbreeding systems will not evolve unless the practical restrictions set by the local production systems are very carefully taken into account.

Table 1: Relative costs and profits derived from crossbred cattle of intermediate levels of European breed inheritance, compared with others under different systems in Latin America (50% European crosses = 100).

Country/SystemBreed GroupRelative
Pasture + supplementLocal cross100100
European100- 300
ConfinedEuropean210- 300
Low level management50% European-zebu100100
25% European-zebu4040
European- 13- 13
High level management50% European-zebu51 
25% European-zebu- 11 
Lowland, medium level management50% European-zebu100100
75% European-zebu10983
Highland, high level managementEuropean16770

* Wilkins et al., 1979
** Madalena, 1989
*** Holmann, 1990

The more common of these practical restrictions deserve further examination. Systems which require the simultaneous use of more than one breed of bull per farm are difficult to put into practice unless there is sufficient infrastructure (fences, records) to separate one bull and his appropriate mates from the other. Although these difficulties could be overcome by changing the bull breed after a period of time, the problem of identifying sources of bulls of known genetic quality at reasonable prices remains. Extending the system to three or more breeds would seem to preclude its use on a wide scale in most practical circumstances. Besides, whatever advantages might be expected from additional heterosis are only of interest if there are no important additive genetic differences between the breeds available for the programme. Thus, for example, rotational crossbreeding schemes involving Zebu and European breeds for milk must take into account the much lower yield to be expected from the Zebu bulls' daughters. Similarly, under Latin American conditions at least, it would be questionable whether a second European breed, besides the Holstein, would make a sufficient contribution to a dual-purpose crossing scheme involving zebu cattle to warrant the additional complication, due to the superiority of the Holstein with it most common rival the Brown Swiss (Syrstad, 1985). it is therefore not surprising, that the evidence available to McDowell (1985) suggested that no advantage has generally been observed from the addition of a third breed to crossbreeding schemes for milk production.

The practical difficulties involved in dealing with various breeds of bull on the farm could in theory be solved by using AI, although the difficulty of obtaining semen of good genetic quality remains. In this context, it is necessary to draw attention to the risks which are commonly associated with using AI for the production of routine pregnancies (as opposed to occasional, strategic uses). Before proposing any crossbreeding scheme which depends on AI, care must be taken to avoid the risk of losing more calves and lactations through lower conception rates compared with natural service, than can be compensated for by genetic improvement. This risk is typically high in lowland tropical cattle populations. From commercial dual purpose herds in Venezuela, Gonzáles (1981) reported a 12% lower pregnancy rate using AI than with natural service and even under experiment station conditions, Paterson et al. (1983) reported a 22% difference between the two mating systems in South Africa. Differences of this magnitude are difficult to justify on the grounds of the genetic quality of the AI sires.

A second practical restriction refers to the lack of infrastructure for selecing local cows and bulls, which affects the potential success of inter se crossbreeding systems and of new “synthetic” breeds. Extremely few lowland tropical communities can take on successfully the challenge of being self-sufficient for genetic improvement, which is required once the population is closed. This probably explains why so few of the attempts to form new breeds in the tropics have succeeded. Cow selection requires more than recording schemes: data must be processed routinely to evaluate genetic merit. To this author's knowledge, no lowland tropical cow population is at present evaluated routinely for estimated genetic merit for production traits. The selection of bulls, through progeny testing, raises practical problems which will be discussed later, although it is relevant to point out that the problems associated with effective bull progeny testing represent one of the limitations of the inter se crossbreeding schemes.

Inter se crossbreeding programmes can also be criticised on the grounds that they are inflexible. Under Latin american conditions, a very small geographical area will include farms with widely different environmental conditions. Which would require animals of different levels of European breed blood and a fixed genotype is therefore of limited use.

These considerations lead to the conclusion that the most widely applicable crossbreeding schemes will allow: a) the use of natural service, b) the simultaneous use of one, or at most two, genotypes of bull per farm, c) the routine introduction of improved germplasm from outside, to bolster local selection efforts or even supplant them under extremely difficult circumstances, and d) the provision of different genotypes so that farmers, with suitable technical assistance, they may generate the type of crossbred which will be most productive under their specific conditions.

An option which may be of wider use is that of the using crossbred bulls, bred from selected local dams and proven European breed sires. Work is in progress to determine the optimum genetic merit of European sires to be used and, once this is clear, the possibility of giving tropical cattle populations access to the huge range of constantly improving temperate zone germplasm will open back. Local selection efforts to identify dams of potential bulls are required, but this can be done in pilot or nucleus herds without necessarily having to carry out recording and selection on the majority of commercial farms. It is assumed that a native (zebu or criollo) populations will be continuously available to provide bull dams using, perhaps, more marginal lands unsuitable for the crossbreds. If not, the scheme degenerates into a “grading up” programme.


It is difficult to demonstrate whether the genetic progress to be expected from selection within an indigenous population is likely to justify the investment required. Heritabilities are seldom known with precision and selection intensity is frequently low under precarious economic conditions since sales tend to be determined more by immediate monetary needs than by biological criteria. On the other hand, the cost of obtaining, processing and interpreting records is likely to be high, because of poor and unreliable infrastructure, and equally difficult to predict.

Despite this, it seems unacceptable to propose that nothing should be done. Tropical cattle herds are extremely variable in production characters (Table 2) and whatever genetic basis for differences between individuals may be, effective management requires that superior cows should be retained and inferior ones culled. The need to get rid of unproductive animals is all the greater under conditions of restricted feed supply. Furthermore, farmers, certainly in Latin America, take great pride in the acquisition of bulls and regularly assign resources to this purpose. The opportunity for introducing improved germplasm into herds by this means should not be missed. Finally, in our experience, the process of encouraging farmers to assess their cattle individually and to take decisions as to their merit, is very well received and associated with spontaneous efforts on their part to introduce other improvements.

Table 2: The potential for selection in commercial dual purpose herds:the variation in production levels between the best and worst cows on five farms in Falcón, Venezuela

Deviation from the group mean (kg)*
Farm NoMilk yield/lactation4 month calf weight

* 1584–2896 kg milk; 58-100 kg calf weight at 4 months.

Selection criteria and methods should be allowed to vary in complexity according to the stage of development of the programme. The important point is to start the scheme on sound, sufficiently simple principles, for it to be carried out properly. A step-wise approach for dual purpose cattle in Latin America which takes this evolution into account has been described elsewhere (Vaccaro and Vaccaro, 1989). At first, cow selection could be based on calf growth and/or milk and reproductive efficiency. The inclusion of some measure of fertility seems essential because of the evidence in zebu populations that heritability is moderate (Plasse, 1988) and the consistent evidence of a negative phenotypic correlation between fertility and milk production under tropical conditions (Table 3). Where herd size is small, approximately valid contemporary groups can be made by uniting data across herds, classified according to production system and, possibly, mean production levels.

Table 3: Example of the negative relationship shown between milk yield and fertility in crossbred European x zebu cows in dual purpose systems in the lowland tropics of Venezuela.

Milk yield/lactation (kg)Interval from calving to conceptionAnoestrus cows (%)
< 100068.117
1001 – 150092.622
1501 – 2000104.728
2001 – 2500121.642
2501 – 3000137.457
> 3000141.965

Source: González (1980)

Bull selection for milk production poses special problems; to assume that they can be effectively evaluated by progeny testing may well be unrealistic. Tropical herds are characterized by low reproductive efficiency, late age at first calving and high mortalities, low culling and lack of identification. In addition, milk yield which is the trait of principal interest is extremely variable compared with temperate zone standards. The size of the population which is mated by AI and performance recorded is usually small in the lowland tropics and, because of the high variation, low fertility and high rates of loss; more dams must be inseminated to produce the necessary number of daughters. This has been estimated at 30 in Cuba Menéndez, (1985) and McDowell (1983) showed clearly the inaccuracies which result when daughter groups are smaller. As a result, too few bulls can be tested reliably to permit an intensity of selection sufficient to justify the whole operation. In addition, the generation interval is prolonged by the late age for production of freezable semen and of first calving, as well as by the typical long process involved in processing and publishing the results. Once superior bulls are identified, their impact on the whole population is limited by the scope of the AI programme and low fertility associated with artificial breeding and by the high rates of loss before the end of the first lactation. An exercise to demonstrate the relative genetic and economic benefits of various selection options for dual purpose herds in Latin America showed little increase in genetic progress for milk due to progeny testing and a reduction in progress for 18 month body weight (Table 4). It was concluded that the exorbitant cost of progeny testing under these conditions could not be justified.

Table 4: Simulated effects of different sire selection options on genetic progress in dual purpose herds under lowland tropical conditions in Latin America

Selection MethodMean Generation Interval (yrs)Cost/bull selected
Genetic Progress per year (kg)
Milk18 mth body weight
Dam records (DR)5.4-8520.72.4
Sire records (SR)5.4-8556.82.4
DR + SR5.4-8576.82.4
DR + SR with embryo transplants5.91,5008587.42.2
DR + SR with progeny testing of bull8.8341,6006589.51.5

Source: Vaccaro (1988)

Practical improvement programmes should therefore explore alternatives to progeny testing. In native populations, emphasis can be given to bull dam selection and in crossbred populations to sire selection as well if, as proposed above, bull sires are routinely selected from temperate zone breeds and used on native dams.

Where progeny testing can be carried out effectively, BLUP procedures are usually recommended. Few tropical populations are likely to have the required information and it is of interest to note that a comparison made in Cuba showed no difference in the ranking of 20 Holstein bulls whether the methods used were BLUP (with or without taking the between bull relationships into account), contemporary comparisons or least squares (Cordovi et al., 1983).

Field performance recording is generally difficult under lowland tropical conditions, especially where herds are far apart and rains seasonally intense. Whole sections of various Latin American countries, including Peru and Colombia, are presently intransitable because of high personal safety risks. The best investment is probably to show farmers how to keep and use their own records. Wives and children will often spontaneously undertake record keeping and this source of enthusiasm should be tapped. Record processing may best be done locally. National organisations usually lack the stability and agility to be effective. Governments should therefore be encouraged to participate in the establishment of overall guidelines for record keeping and then delegate responsibility to properly trained scientists working locally in institutions such as universities. The schemes should start on a small scale. The most usual bottlenecks are the time lag between record collection and return of the processed information to the farm, and the failure to present the results in a way which permits farmers to see the order of genetic merit of their animals for traits of priority importance. These common faults should be overcome before the programme is allowed to expand.

Given the difficulties set out above the establishment of pilot or nucleus herds appears essential. In extreme cases, recording and selection would be confined to these farms. The effects would filter down into the rest of the population at rates which vary according to their number and size.

Nucleus herds would generally be considered to be specific units but a very simple low cost scheme consists of providing suitable scientific advice to progressive commercial farms and using them as nucleus herds to provide at least some of the bulls and replacement females required by the rest. Care must be taken to admit only those farms which represent the priority production system, avoiding the temptation to use high yielding herds which do not. Also, the farmers must be genuinely convinced that the selection criteria agreed upon are valid. The applicability of the proposal will vary between regions and will depend partly on herd size. The important principle is that official organisations in developing countries tend to be unstable and poorly budgeted, and there are clear advantages in locating the programme directly in farmers' hands. One of the most important of the advantages is that it minimises the risk of selection under unrepresentative environmental conditions. The programme should then grow under its own momentum and be sustainable in the sends that it becomes increasingly independent of outside inputs. Such a scheme is currently in progress in Venezuela. Outstanding zebu cows are inseminated with semen from proven Holstein bulls and the young males distributed to surrounding farmers, who are kept informed of the number, genetic credentials and prices of the bulls available. The university base of the project is useful, at least in the first years of the programme, to give some assurance to purchasers that information is unbiased.

In other circumstances, it may be necessary to collect animals into one site to that proper recording and selection can be undertaken. For dairy cattle (or sheep) it might be the only practical way in which bull mothers can be mated to a few selected sires. If such a unit can be established on the basis of reasonably accurate estimation of breeding value, the genetic lift can be substantial. However, this may not be possible in all cases, and the major gain could simply be having adequate numbers to provide contemporary comparisons and the opportunity of recording objectively.

Screening a population to obtain the “best” animals is an important component of the establishment of a nucleus. Even where an existing herd is used as a nucleus, it is useful to sample the population outside albeit with the objective of only taking very few animals in. The principle of using contemporary comparisons can be maintained even though the group may be a village rather than each villager's herd.

There are clearly some criteria which can be used in all circumstances - contemporaneity and acceptability on a physical basis (if this is not so, dissemination may not take place). The criteria will range from proper direct measurements to stockmen's memory/judgement. If possible, it is useful in the early stages to obtain average animals from the same sources as the outstanding ones. Once together, the Comparison will provide a useful guide to the accuracy of the criteria and a direct indication to those interested of the value of genetic selection. Examples of this type of screening and nucleus formation are given by Timon (1990) and the results shown in Table 5.

Table 5: Genetic Screening Results - Awassi Sheep Turkey

Lactation Yield (kg)
310   7.7
(254 – 469)
223   9.3
(97 – 360)
+ 3%
Lactation Length (days)
206   1.7
(159 – 224)
187   4.4
(95 – 222)
+ 10
Maximum Daily Yield
2.7   0.1
(2.1 – 4.2)
2.1   0.1
(1.0 – 4.3)
+ 29%
No of animals4343 

Nucleus herds can allow selection to take place with limited training (in terms of numbers of people) since only the staff needed to run the unit and to record the necessary parameters need be trained. As long as proper monitoring is built in to such a system there is probably little need for a geneticist on site (except as a regular but infrequent visitor). While a central nucleus enables additional records to be taken (if cost effective), there is always the risk that the environment may not be representative of the sustainable production system or of a reasonable level of intensity within that system. The nucleus should never be provided with an environment better than that anticipated for commercial production two/three generations ahead. At least, such a constraint should allow the correlated response in commercial production to be within acceptable limits.

Open nucleus breeding schemes (usually known as ONBS) under most circumstances can achieve faster rates of genetic change although under certain circumstances (MLC, 1981) progress can be lower. However, the real advantages of the ONBS are the reduction in inbreeding and the fact that dissemination of the improved genetic material is built into the system. Such a fact also provides a good reason for achieving high health status in the nucleus but this can lead to major problems as stock transferred down the pyramid may well not be able to withstand the health challenges encountered.


Perhaps one of the problems of all schemes in developing countries is the fact that if infrastructure is poor for recording it is equally poor for dissemination of genetic material. Certainly the derelict European style AI centres in Africa are witness to the problem. However, effective dissemination can be achieved using local resources as exemplified by the use of pig AI in some areas of China where local public transport and bicycles are the main forms of transport. The problem of heat detection is always present where herds are of few animals and with buffalo, the problem is present in all herds unless a bull is used for this purpose. In cattle, the risk involved in the use of AI to obtain routine pregnancies under most lowland tropical conditions has been pointed out above. Usually, therefore, AI should be confined to use in specific cases, of which bull breeding would be one and the production of an initial crossbred generation perhaps another. Where natural service is required to ensure adequate birth rates, a great deal can be done to help farmers organise a reliable supply of males of known genetic quality through cooperatives or more informal networks.

Large quantities of resources are currently spent even in poor tropical countries on MOET. The excitement of new technology easily diverts scientists' (and politicians') attention from the strict socio-economic objective of the production system. The possible benefit of alternative uses of the same resources (e.g. through AI) must be carefully measured before MOET is accepted as a viable option. The true genetic merit of the donor females must also be properly evaluated before transfer is carried out. The difficulty of doing this properly should not be underestimated, especially for traits such as milk production which are of low heritability. It is possible that ET and also cloning could provide useful methods of multiplication. Where AI is likely to be difficult, it may be more feasible to provide “improved” male embryos for rearing within the locality so that distribution is achieved while still relying on natural service. Whether implantation is done centrally and females distributed or is done in the locality will depend on local circumstances. Again, however, it is essential to estimate the cost of the operation in terms of genetic gain and determine whether it does indeed make best possible use of existing resources.


It is widely accepted that genetic diversity is an essential element of the long term sustainability of production systems. It seems, however, unrealistic to propose that countries which now face severe problems of food production and poverty should devote their resources to the preservation of genotypes which are not at present commercially viable. Reduction in population numbers usually means that the economic potential of the animals is uncompetitive. It could also be argued that, given the wide diversity of natural ecosystems throughout the temperate and tropical world, a diversity of genotypes will automatically be maintained in production systems, at least in the case of species which are commonly relatively little protected from the natural environment (e.g. ruminants). In that case, it should be possible to obtain genes which might be required at a given moment, from populations which are maintained commercially in some ecosystems, even perhaps on another continent.

From these considerations it would appear, first, that endangered breeds in developing countries should not necessarily all be conserved. Realistic criteria must be established for deciding which cases are justifiable. These should include productive and adaptive aspects and the process could possibly be refined by genetic distance estimation procedures. It would also seem reasonable to propose that funds for this purpose should be obtained from international sources and that the work should be organized on a regional basis so that complementarity between individual nations' efforts can be improved. In considering conservation methods, the possibility of maintaining populations in natural parks or reserves should not be discounted although the risks of loss from disease and hunting must not be underestimated. Finally, governments could reasonably be encouraged to take concrete measures to evaluate existing, small populations which could prove useful in the future in specific circumstances (e.g. in crossbreeding schemes or in traditional, small-farm production systems). One important element is to carry out the local research required to demonstrate the value of such animals in production systems considered most likely to be sustainable.


While it is relatively simple to comment with hindsight on programmes, the initial planning and development of schemes is frequently difficult. The difficulties often stem from the fact that different pressures are put on those involved in planning improvement programmes - political (governmental), breed societies/cooperatives, regional interests - and, in general, a belief that breed improvement is an automatic and acceptable solution. Since genetic improvement is relatively slow to provide change it can never be the “quick fix” so often cherished by politicians. The failures of so many schemes based on exotic semen are witness to that fact. The planner requires information on:

Only then can long term, sustainable systems be planned and specific projects within the plan can be selected on the priorities agreed by those in the decision making role. This provides a more useful background for improvement since specific projects are known to fit into an overall picture rather than the more frequent occurrence of adopting a project and then trying to fit it into the long term strategy.


Cordovi, J., Guerra, D and Menéndez A., 1983. Comparación de métodos utilizados en la evaluación de toros para la producción de leche. Centro de Investigación para el Mejoramiento Animal. La ana, Cuba. Mimeo, 7 pp.

González, C. 1980. Report of Cattle Reproduction Group. Universidad del Zulia, Maracaibo, Venez la.

González, C. 1981. Factores que a-fectan la fertilidad al primer servicio en vacasestizas. VIII Meeting, Asociación Latinoamericana de Producción Animal, Dominican Republic; F 21 (Abstract).

Cunningham, E.P. and Syrstad, O. 1987. Crossbreeding bos indicus and bos taurus for milk production in the tropics. FAO Animal Production and Health Paper 68. Rome, Italy.

Holmann, F. 1990. Grupos Genéticos y Sistemas de Producción de Leche en Paises Tropicales: Experiencias en Investigación y Programas de Desarrollo. International Development Research Centre, Ottawa. 89 pp (in press).

Madalena, F. 1989. Cattle breed resource utilization for dairy production in Brazil. Revista Brasileira de Genética 12: (Supplement) 1983–220.

McDowell, R.E. 1983. Strategy for improving beef and dairy cattle in the tropics. Cornell International Agriculture. Mimeograph No. 100. Cornell university, Ithaca, N.Y. 34 pp.

McDowell, R.E. 1985. Crossbreeding in tropical areas with emphasis on milk, health and fitness. Journal of Dairy Science 68: 2418–2435.

Menéndez, A. 1985. Breve reseña sobre la productividad del vacuno en Cuba. II Jornadas sobre Ganadería de Doble Propósito, Maracaibo, Venezuela. Mimeo, 8 pp.

Minson, D.J., 1980. Nutritional differences between tropical and temperate pastures. In: Morley, F.H.W. (Ed.) Grazing Animals. Elsevier, Amsterdam. pp. 103–157.

MLC. 1981. Meat and Livestock Commission 1981. Group Breeding Schemes for beef cattle. Authors Steane, D.E., Guy, D.R. and Smith, C.

Plasse, D. 1988. IV Cursillo sobre Bovinos de Carne. Ed. Plasse, D. and N. Peña. Facultad de Ciencias Veterinarias, Universidad Central de Venezuela, Maracay.

Preston, T.R. 1976. Prospects for the intensification of cattle production in developing countries. In: Smith, A.J. (Ed.) Beef Cattle Production in Developing Countries. Univ. Edinburgh Press, Scotland.pp. 242–257.

Preston, T.R. and Leng, R.A. 1987. Matching Ruminant Production Systems with Available Resources. Penambul Books, Armidale. 245 pp.

Seré. C. and Jarvis L, 1989. The betting line on beef: ex-ante estimates of improved pasture research benefits for the Latin American tropics. Trends in CIAT Commodities. Centro Internacional de Agricultura Tropical, Cali. pp. 68–88.

Seré, C. and Vaccaro, L. 1985. Milk production from dual purpose systems in tropical Latin America. In: Smith, A.J. (Ed.) Milk Production in Developing Countries. Univ. Edinburgh Press, Scotlnd. pp. 459–475.

Syrstad, O. 1985. Relative merits of various Bos taurus dairy breeds for crossing with Bos indicus cattle. Livestock Production Science 9: 86–94

Timon, V.M. 1990. Genetic Improvement of Small Ruminant populations in developing countries. Proc. 1st World International Symposium on Sheep Production, Zagreb, 1990.

Vaccaro, L. 1990. Survival of European dairy breeds and their crosses with zebus in the tropics. Animal Breeding Abstracts 58: 475–494.

Vaccaro, L. and Vaccaro, R., 1989. Guidelines for the genetic evaluation of dairy and dual purpose cattle in the Latin American tropics. Revista Brasileira de Genética 12 (Supplement): 27–35.

Vaccaro, R. 1988. Metodología apropiada en la evaluación del valor genético de animales en poblaciones cruzadas de doble propósito. Ciencia Tecnol. Venezuela 5: 73–82.

Wilkins, J.V., Ali and Vaca Diez, C. 1979. El cruzamiento para la producción de leche en los llanos bolivianos. VII Meeting, Asociación Latinoamericana de Producción Animal, Panamá, 7 pp.

Winrock International, 1978. The role of ruminants in support of man. Winrock International Research and Training Center, Arkansas, USA. 136 pp.

Previous Page Top of Page Next Page