Fourteen years ago, in 1976, at the third international Symposium on the Preservation of the Przewalski horse, we expressed our serious concern for the long-term survival of the captive population of these animals; at the fourth Symposium in 1980, we did repeat our deep anxiety. At that time the demographic and genetic analyses of BOUMAN (1977) and of BOUMAN & BOS (1979) had clearly shown, that consanguineous matings have produced generations of increasingly inbred horses (RYDER et al. 1982).

Higher juvenile mortality and decreased life span were signs of a reduction in fitness in successive generations. The natural reservoir of genetic variability which constitutes the gene pool of the species, had been diminished by bottlenecks in the breeding history and by management procedures such as breeding in pairs, or small groups, selection, inbreeding and over-employment of certain stallions. The analyses of the problems in the breeding history of the Przewalski horses clearly pointed to the fact that in captivity the populations had changed and experienced a different evolutionary development than it would have had in the wild.

Since than we have tried to stimulate international cooperation between breeders of Przewalski horses. Our data system, from which it is easy to extract all data on any one horse as well on a line of horses, was made available in order to analyse their breeding potential. In cooperation with Dr Jiri Volf, the former keeper of the Przewalski horse studbook in Prague, a bulletin was sent twice a year to each individual collection with data on the new born foals as well as additional information on inbreeding coefficient and founder representation, all indicated by various colour codes. With the bulletins we hoped to bring about a change of heart, resulting in the sense to use less interrelated Przewalski horses for breeding purposes and/or to acquire horses from different genetic backgrounds (DATHE 1980).

The Arnhem Study Conference on genetics and hereditary diseases of the Przewalski horses was organised in 1978 by the Foundation for the Preservation and Protection of the Przewalski Horse (FPPPH) in cooperation with the Arnhem Zoo to exchange knowledge and stimulate further cooperation in research. This resulted in the setting up of a working group on diseases of the Przewalski horses (ASTHON 1979 and 1984). Unfortunately time was then not ready for international cooperation, in any case not in Europe. We are happy therefore that coordinated and cooperative breeding plans for the Przewalski horses in zoos got off the ground, such as the SSP (Species Survival Programmes) established in 1979 in the USA and the EEP (European Endangered  Species Programmes) later, in 1986 (RYDER et al. 1982). We were astonished, however, that the EEP could not find a solution to give the Foundation Reserves for the Przewalski horse, which owns a large herd of healthy horses in the semireserves in the Netherlands, a chance to become a member.

Our main interest has always been and still is the reintroduction of the Przewalski horse into the wild.

The analyses based on our studbook inventory card system also convinced us however that a totally different management strategy is needed for a wild animal species which is to be reintroduced into natural habitat than for those individuals which are destined for permanent captive breeding in zoos. It is not only important to maintain the genetic variability of these species by cooperative breeding plans, but they have to be kept in species specific environments where they can develop a normal social behaviour repertoire, where selective pressures can be more natural and the duration of captive breeding is minimised.

When we analyse the breeding of the Przewalski horses in zoos and private parks the process of creeping domestication is evident. Under longterm captive conditions other phenotypic and behaviour genotypes came to expression due to unnatural selective pressures caused by captive environment, human selection and human interference in the normal patterns of mate choice and of dispersal of adults and young (HERRE 1967, BOUMAN 1989, VOLF 1989). These irreversible changes in the genepool cannot be nullified.

Practically no other wild species has yet been kept in captivity for so many generations with almost no new genetic input. We expect that the Przewalski horse, when released into a presumed natural habitat, will suffer enormous losses. The selective pressure will even be stronger because many wolves will be around in Central Asia and Mongolia. These thoughts motivated us years ago to develop a plan to enlarge their change for survival after release, by trying to increase their fitness and their adaptability to changed environments in semi-reserves before planning their release into the wild. We discussed the plan with Sir Peter Scott, then chairman of the Survival Service Commission of the IUCN, and in 1979 he asked us to set up a Captive Breeding Group for the Przewalski horse.

A variety of experts in wildlife management, behaviour research and captive breeding were invited to participate in a meeting in the fourth Symposium for the Preservation of the Przewalski Horse in Winchester in 1980. The Foundation for the Preservation and Protection of the Przewalski Horse (FPPPH) proposed a plan for re-introduction into a wild reserve based on the principle that Przewalski  horses cannot be released directly from zoos into the wild.

An intermediary phase in semi-reserves is essential. The necessity of establishing semi-reserves for the Przewalski horse was discussed, accepted and later endorsed by the Symposium (BOUMAN 1979, DATHE et al. 1984). In the same year the Foundation Reserves for the Przewalski Horse came into being in the Netherlands, a joint enterprise between the FPPPH and the WWF-Netherlands, aiming at the promotion of semi-reserves (IUCN/WWF 3077). In 1990 a sister Foundation “Deutsche Stiftung Urwildpferd” was established.  Five large semi-wild reserves existed in the Netherlands and Germany. Together 56 Przewalski horses of a variety of bloodlines and little related with each other have been released divided in breeding groups or bachelor group. In the semi-reserves the Przewalski horse get the chance to adapt themselves to a semi-natural living in large territories varying in size from 30 to 250 ha.(75-625 acres), where they had to search for their own food and could raise their offspring in more natural social relations. Although a minimum of human contact and interference is aimed at, the horses still need to be controlled and watched for behavioural or genetic problems.

It is important however that selective pressures have to be made more natural, weak foals have to die and horses with genetic diseases are rejected. Only the best “quality” of Przewalski horses must get the chance to be selected for future release. This intermediate stage may probably have to last more then one generation, dependent of the results of running semi-wild before they can be released in wild reserves.