Previous Page Table of Contents Next Page

The conservation of animal genetic resources in the developing countries: a practical way forward

by F. Madalena


Important economic gains may accrue from the appropriate choice of the livestock genetic resources correctly utilized in a given production system. This is illustrated by results a trial on dairy cattle crossbreeding strategies undertaken in Brazil, in which the accumulated lifetime performance records of 527 females have been recorded over the past 14 years on 67 farms. Six levels of crossbreeding between Holstein-Friesian (HF) and zebu Guzera (Z) were included with the HF genetic composition ranging from 1/4 to ≥ 31/32 HF fractions. A summary for the first 9 years is in Table 1.

Farms were grouped into two management classes for analytical purposes. The F1's were the most profitable group under both management regimes and for this reason was taken as a reference point to express relative performance of the other crossbreeding strategies.

Rotational crossing of HF sires for two generations followed by one generation of zebu sires (HF-HF-Z) was the second best alternative in the high management group. Upgrading to HF was as profitable as the HF-HF-Z rotation under high management, except under the higher fat and protein pricing which would be fairer to farmers. However in the low management grouping, upgrading to HF would have had disastrous consequences due to low production, high mortality and low culling rates. A new breed developed from inter se matings would require a high selection intensity to reach the same profitability as the F1. Choice of strategy would, of course, depend not only on performance but also on the actual possibilities of implementing each crossbreeding scheme (Madalena, 1989).

The above example shows that by using the right type of animal, without changes in either nutrition, health or other inputs, profit was considerably increased. Conversely, losses resulted from the inappropriate choice of crossbred. This, of course, does not mean that environmental factors should not be improved, but rather that both the genetic and environmental components should be considered in unison. The choice of germplasm is an integral element in the production system and it must be carefully matched to the other available inputs.

It might be appropriated to discuss the implication of the genotype x environmental interactions that are seen in the Brazilian trial. The view is sometimes expressed that there is no need to worry about genetics until the management is sufficiently improved to allow full expression of the existing available genetic potential. This view, however, fails to recognize that some genotypes have higher potential in a favourable environment but a lower potential in more stressful conditions. Therefore the notion that there is a genetic potential for each level of management is conceptually and practically more accurate (Falconer, 1960). As illustrated by the results shown in Table 1 ignoring genetic differences would be an unwise decision in any development programme.

Table 1: Profit per cow per day of herd-life under alternative strategies of crossbreeding of Holstein-Friesian (HF) x zebu (Z).

 Management Level*
F1 profit***
Strategypercent of F1 performance
HF-HF-Z rotation7572848148515247
HF-Z rotation4153635759606163
Upgrading to HF75578480-21-12-10-28
New breed

* Characteristics of management levels:“High”“Low”
Mean first lactation milk yield (kg)24501731
Mean first lactation length (days)283309
Mean first calving interval (days)402561
Concentrates fed, kg/cow/day4.51.6
All milked 2x/day with calf suckling to stimulate let-down.

** A: 1980 to 1985 prices (protein not paid for)
B: Fat differential tripled and protein paid at same rate.
C: Cost of concentrates halved.
D: Beef value of animals doubled.

*** Profit expressed in kg milk/day.
Price of 1 kg milk (3.3% fat) = US$ 0.16.

Source: Madalena et al. (1989).


Historically, man has reshuffled genes from different livestock populations by crossing, selection and inbreeding. Biotechnology now offers powerful new methods to change the genetic composition of animals. However, because genetic material cannot be synthesized, improvements will still be restricted to the obtaining of the best possible combinations of existing DNA. Therefore, animal genetic resources constitute an indispensable natural resource must be properly managed for efficient production now and to be preserved for future use.

Although germplasm introduction has traditionally been a common practice in animal production, it was only really in the last two decades that genetic variation came to be viewed as a natural resource (Dickerson, 1969). Commercial use of genetic variation between populations has increased dramatically in some species as has research, primarily in developed countries, regarding the its theoretical aspects and the evaluation of alternative breeding plans. Although migration also plays an important role in the genetics of animal populations in developing countries, research generally did not proceed, or even accompany, commercial trends. Genetic resources in these countries have not been adequately evaluated or fully utilized and in some cases are threatened with extinction without being properly described. The following activities might be listed for action on genetic resources management:

FAO's programme has recently been described by Hodges (1990) and I will briefly refer to the above in the following sections, concentrating on issues rather than methodology.


A brief description of the world's main livestock species and breeds is given in Mason's (1988) “World Dictionary of Livestock Breeds”. A Global Data Bank of Animal Genetic resources has been established jointly by the European Association of Animal Production and FAO at the Institute of Animal Breeding and Genetics, Hanover School of Veterinary Medicine (Simon, 1990). Information from the bank is publicly available. A questionnaire was developed to collect information on breed origin, numbers, phenotypic description, uses, management systems and relative performance compared to standard breeds. The work was initiated with cattle, buffalo, sheep, goats, pigs and horses, however, questionnaires for poultry and Andean cameloids are in preparation. The databank currently has information on 658 breeds from 25 countries (mostly European, USSR and China) and procurement of data from other countries is being sought. In many cases such information is locally available although it is not generally readily accessible at hand, so compiling it in one databank requires motivation and effort, especially for completing the questionnaires. In other cases, special field surveys may be required to obtain the relevant information.

Estimates of genetic distance between breeds would be valuable in formulating conservation policies, thereby allowing a more rational selection of breeds requiring preservation. Where possible DNA description is preferable to indirect description involving gene expression (Primard, 1985). Trinity College, Dublin, is undertaking a study to measure genetic distances between 6 european and 8 tropical breeds (from India and Africa) of cattle. DNA finger-printing based on endonuclease restriction length fragments is being evaluated and both mitochondrial and nuclear DNA concentrates are being extracted locally and transferred to Dublin for analysis (D. McHugh and R. Loftus, personal communication).


The systematic introduction and evaluation of germplasm is a common practice in plant breeding but less so in animal breeding, except perhaps in poultry. However, important work has been undertaken such as the trials established by the Meat Animal Research Centre in Lincoln, Nebraska, where a large number of cattle breeds have been evaluated over the past 20 years; the early Argentinean beef breeds comparison trial in the 1960's; the evaluation of European breeds in Great Britain and the FAO trial in Poland to compare Holstein-Friesian strains from 10 different regions.

The evaluation of germplasm is just one particular aspect of the overall research requirements. One hesitates to emphasize germplasm evaluation since the benefits would appear to be obvious. Yet, there have been many cases where recommendations based on opinion, rather than of results, have had disastrous effects. This is particularly the case where livestock importations have been involved and Vaccaro (1990) indicated that european dairy breeds are not sustainable. Also the general belief that 5/8 European x 3/8 zebu cattle crosses were the best combination for tropical environments has caused unnecessary delays in new breed developments (Madalena, 1989) and, in spite of early warnings (eg. McDowell, 1972), it has only been possible to put the matter in its proper prospective in the light of recent experimental results.

Another example of the importance of germplasm evaluation may be found in the Criollo, cattle that were originally introduced into Latin America, and that have adapted over the centuries. The Criollo have now almost completely subsumed by the zebu in the lowland tropical areas, a process that intensified in the 1920's (de Alba, 1987). Recent research has shown that purebred zebus are no more productive than the purebred Criollos, although the F1 crosses are superior to both, due largely to heterosis expressed in reproductive efficiency and other economically important traits (Plasse, 1989). Therefore, a crossbreeding policy would have been indicated instead of the breed substitution. As de Alba (1987) pointed out, farmers and ranchers impressed by the crossbred's performance attributed it to the new breed and not to heterosis - which is a more difficult phenomenon to grasp. The result was that a population of over 100 million head were pushed in the wrong genetic direction. Had the research been done 70 years ago and the results explained to farmers, this situation might have been avoided.

To be useful, germplasm evaluation must be comparative. The breeding alternatives should be compared over the same environments using appropriate designs, which include sufficient numbers of representative animals and sires. Lifetime performance needs to be recorded, since both survival and herdlife are major components of the overall economic performance of breeds in stressful environments, and large differences in these traits are expected for diverging genotypes. With ruminants, this means that evaluations have to be in the order of 10 years. Shortcuts are not available and would only lead to underestimation of the real economic differences between alternatives.

Germplasm evaluation should, wherever possible, be conducted under commercial management rather than at experimental stations. In developing countries government farms are often inadequately funded, poorly administered and it may not be easy to simulate the variety of management and socio-economic situations found in the private sector. For example, it is unlikely that administrators would allow the poor husbandry conditions found in some commercial farms. While the basic components of economic performance should be recorded on farms, some investigation of the more sophisticated traits may need to be undertaken to understand why the observed differences occur. For example, product quality may be conveniently analyzed at a central laboratory, although getting the samples to it may be not an easy task.

Operationally, on-farm trials are usually cheaper, since farmers do not carry bureaucratic overheads. It is also less likely that administrative changes would disrupt long-term experiments in farms as can happen in experimental stations. On-farm recording, however, although not expensive, requires organizational and logistical support along with careful supervision.

I submit that the real opportunity cost of on-farm germplasm evaluation is not high. The costs of producing the necessary animals, distribution and performance recording may be largely recovered from production through some form of share-farming schemes as was done in the Brazilian trial. Staff and facilities for supervision, analysis and to interpret results would be available at research institutes/universities in many countries, although they are likely to be overburdened with other activities. Therefore some support would be needed to strengthen laboratories, hire technical staff and provide minor, unsophisticated equipment and reagents.


Characteristics of both species and breeds result from the expression of not only individual genes but also from gene combinations which interact to influence the physiological processes. Genes and gene combinations are renewed by reproduction, however, they may also be lost through either the extinction of the population or replacement by other genes or gene combinations. Selection and crossing will cause genetic change in a given direction, conversely, a low population size and inbreeding may result in the random loss of genetic material. Depleting genetic variation will restrict the choice of available genetic material for use by future generations. Should new circumstances arise, requiring animals with new characteristics then the genetic material would not be available to develop it. Conservation of animal genetic resources must be seen as one aspect of the wider problem of maintaining bio-diversity. Present increased resource utilization conflicts with possible future needs.

As the decision making process becomes increasingly centralized through the development of rural organizations, communications, education and propaganda; and the increasing use of reproduction tools like artificial insemination and embryo transfer, so the power of man to change the genetic make-up of livestock will increase dramatically. Cases are already known of breed replacement on almost a continental scale within a few decades. Some preservation is clearly needed to allow future generations the opportunity to be able to use the genetic resources that are currently available today. Thus, conservation of genetic resources could be considered as an insurance. Studies have shown that the economic return is very high when the preserved germplasm is extensively used (Smith, 1984). However, the likelihood of the preserved resources being useful in the future is, by definition, not known so the decision on how much to spend remains largely a subjective one.

At present, there are two basic ways of preserving genetic material for further utilization, either as live animals or as deep frozen semen or embryos. Freezing of ruminant semen has been used for a long time and embryos for the past decade. Freezing of pig and horse semen is a more recent development, although freezing pig embryos is not yet possible. Freezing of poultry semen is not yet commercially feasible. The conservation, or reactivation, of other cells, chromosomes or DNA should still be considered at an experimental stage (Brem and Brenig, 1990). Because biotechnology is moving so fast, techniques may rapidly evolve, so it may be appropriate to set a conservation planning horizon of, say, 25 years. Future users would then decide if and how to reactivate stored material; continue to preserve it (or indeed to throw it away!) in the light of available techniques at the time. A short planning horizon is not reasonable, because most of the conservation cost is incurred initially, to establish herds or to freeze the semen/embryos.

Judgement is required to compromise between the size of the genetic sample, the number of unit to be preserved and the cost of preservation cost. Based on inbreeding considerations Smith (1984) suggested storing 100 semen doses each from 25 unrelated sires and 25 embryos from each of 25 unrelated females donors and sires per breed. Indicative costs of semen preservation for 25 years would be US$126,000 per breed and US$158,000 for embryo preservation. Sixty-five and 86 percent of these costs, respectively, correspond to the initial collection cost while maintenance costs are relatively inexpensive. Although semen is cheaper to preserve, reactivation would be faster if embryos were also available.

Alternatively, a herd/flock of 25 breeding females could be kept with no inbreeding over a 25 year period, provided controlled matings or artificial insemination was possible. Under pasture management this should not be expensive. Thus, in situ, conservation appears to be simpler than cryo-preservation. However, the risks of losing genetic material due to constraints in population size are higher for live animals. Disease outbreaks, droughts, broken fences, periods of poor management and the intrusion of unwanted sires in the herd might not be rare occurrences and would need to be guarded against.

Therefore, each conservation method has its merits and weaknesses which need to be assessed for each particular situation. For example, some countries have established embryo manipulation facilities and these could be made available to other countries in the region, such as, the proposal for Regional Gene Banks promoted by FAO.

In general, conservation of genetic resources is a long-term activity but there may be special cases in which its objectives may complement shorter-term development goals i.e. in the case of breeds that are commercially under-utilized, in spite of their potential to improve the economic efficiency of present production systems. Yet these valuable breeds may not be preferred by farmers or decision makers for a number of reasons, including: lack of research regarding their potential contribution, propaganda and/or vested interest. Germplasm from these breeds should be evaluated in the context of designing breeding strategies, including breed development, which would lead to their commercial conservation.

Development and conservation must go together. In fact, some contend that under-development leads to loss of natural resources (D. Wood, personal communication). There is also, however, a moral issue: it is difficult to think of saving endangered breeds when confronted with children suffering from hunger caused by famine. Yet, we do not wish to hand over to the next generation a world depleted of the majority of it's natural resources. Essentially mankind has the necessary genetic resources in terms of breeds and individuals to meet the challenge of feeding itself in the future (Cunningham, 1991). Some conservation effort is justified to sustain development in the long term and a balance will have to be established. It is certainly not logical to preserve resources for future use when we do not properly use them now. This would be like spending all your money insuring a car and then not be able to buy the petrol to run it.


Brem, G. and Brenig, G. 1990. Applications of modern cryogenic methods in conservation of animal genetic resources. Proc. 4th World Congr. Genet. Appl. Livest. Prod. XIV: 475–479.

Cunningham, E.P. 1991. Biological Divesity and the Development of the Livestock Industry.

De Alba, J. 1987. Criollo Cattle in Latin America In: Animal Genetic Resources. Strategies for Improved Use and Conservation. FAO Animal Health and Production Paper, 66: p. 19–43.

Dickerson, G. 1969. Experimental approach in utilizing breed resources - Animal Breeding Abstr. 37: 191.

Falconer, D.S. 1960. An Introduction to Quantitative Genetics. Oliver and Boyd, Edinburgh.

Hodges, J. 1990. Animal Genetic Resources. A Decade of Progress. 1980-1990. FAO, Rome.

Madalena, F.E. 1989. Cattle breed resource utilization in Brazil. In: Proc. International Symp. Utilization Animal Genetics Resources in Latin America. Rev. Brazil. Genet. 12 3 Suppl: 183–220.

Madalena, F.E., Teodoro, R.L., Lemos, A.M., Monteiro, J.B.N. and Barbosa, R.T. 1990. Evaluation of strategies for crossbreeding of dairy cattle in Brazil. J. Dairy Sci. 73: 1887-1901.

Mason, I.L. 1988. World Dictionary of Livestock Breeds. C.A.B. International, Oxon, U.K.

McDowell, R.D. 1972. Improvement of Livestock Production in Warm Climates. Freeman, San Francisco.

Plasse, D. 1989. Results from crossbreeding Bos taurus and Bos indicus in tropical Latin America. In Proc. International Symp. Utilization Animal Genetics Resources in Latin-America, Rev. Brazil Genet. 12 3 Suppl. 163–182.

Primard, Catherine. 1985. Divergence morphologique et divergénce moleculaire. Estimation des divergences au niveau des molécules d ADN. In: Les distances génétiques - Estimations et applications, INRA, Paris.

Simon, D. 1990. The Global animal genetic data bank. In: Animal Genetic Resources. A Global Programme for Sustainable Development. FAO. Animal Production Health Paper No. 80, Rome, pp. 153–159.

Smith, C. 1984. Estimated costs of genetic conservation in farm livestock. In: Animal Genetic Resources Conservation by Management, Data Banks and Training. FAO Animal Production and Health Paper 44/1: 21–30.

Vaccaro, L. de 1990. Survival of European dairy breeds and their Crosses with Zebus in the Tropics. Animal Breeding Abstract 58: 475–494.

Previous Page Top of Page Next Page