Botanic Gardens Conservation International
BGCI provides a global voice for all botanic gardens, championing and celebrating their inspiring work. We are the world's largest plant conservation network, open to all. Join us in helping to save the world's threatened plants.

Botanic Garden Seed Banks/Genebanks Worldwide, their Facilities, Collections and Networks

Volume 2 Number 9 - December 1997

Lalibert├ę, Brigitte

The conservation of biodiversity can be achieved through an integrated approach balancing in situ and ex situ conservation strategies. The preservation of species in situ offers all the advantages of allowing natural selection to act which cannot be recreated ex situ. The maintenance of viable and self-sustainable populations of wild species in their natural state represents the ultimate goal, but habitat destruction is inevitable and endangered species need to be preserved before they become extinct. Ex situ conservation can also provide the opportunity to study the biology of, and understand the threats to, endangered species in order to eventually consider successful species recovery programmes which would include restoration and reintroduction. It also has the advantage of preserving plant material and making it available for research purposes, without damaging the natural populations. Their conservation ex situ is therefore complementary to in situ conservation and can act as an "insurance policy" when species are threatened in their natural habitats.

Ex situ Conservation Methods

Ex situ conservation of plant genetic resources can be achieved through different methods such as seed banks, field genebanks, in vitro storage methods, pollen banks and DNA banks. The major consideration for long-term conservation of germplasm collections is the determination of the seed behaviour of each individual species to be preserved during storage under dry conditions and cold temperatures. If the seeds can be dried to a low percentage humidity such as below 8%, in the majority of cases, the seeds will then withstand very cold temperatures of below 20ºC. When the seeds tolerate these conditions and remain viable after many years of storage, they are classified as orthodox (desiccation tolerant) as opposed to recalcitrant (desiccation intolerant) when they do not. However, there might be many other reasons why a particular species cannot be preserved in seed banks, such as a very low production of seed or long-life cycle species, such as trees and perennials. Their conservation as seeds would make the study of their biology difficult.

Seed Banking

Undeniably, the most cost-effective method of providing plant genetic resources for long-term ex situ conservation is through the storage of seeds under very specific conditions, following techniques well developed for crop plants by organisations such as the International Plant Genetic Resources Institute (IPGRI), previously the International Board of Plant Genetic Resources (IBPGR) and the Food and Agricultural Organisation of the United Nations (FAO). The main advantage of seed banking is that it allows large populations to be preserved and genetic erosion to be minimised by providing optimum conditions and reducing the need for regeneration (Given, 1987). However, when a natural population still exists, it may be advisable to re-collect rather than regenerate a new supply from the previous collection as damage can occur such as mutations associated with the loss of viability during storage. The success of long-term conservation of seeds is dependant on continuous viability monitoring and regeneration or re-collection when the viability of the sample drops below a minimum level (Eberhart, Roos & Towill, 1991). It is important to realise that however much care is taken during seed collection, regeneration and storage, natural selection cannot be simulated and some artificial selection will be unavoidable, which inevitably leads to unpredictable genetic changes (Ashton, 1988).

The recommended preferred standards for long-term seed storage of orthodox species recommended by IPGRI is to dry the seeds to a moisture content of below 7% and seal the dried seeds in a moisture-proof container such as laminated foil bags, aluminium cans or glass jars for storage at a low temperature of -18ºC. Clearly this is only applied to true orthodox species of crops and their relatives. However because less is known about wild species, a temperature of -4ºC and a moisture content of 7-8% is advisable to begin with. The activities in seed banks should take the following sequence: collection, seed preparation, seed drying, packaging, storage, periodic germination tests, seed regeneration, re-storage and documentation at each stage of activity.

When plant species are recalcitrant or long-term conservation cannot be achieved through seed banking, different methods have been developed with their respective merits, such as field genebanks, in vitro germplasm collections, pollen and DNA banks. This is in many cases the main problem faced by botanic gardens dealing with many different species of which, a great proportion are be recalcitrant. They need to develop complementary conservation techniques and adopt different methods.

Field Genebanks

Field genebanks or living collections are the main conservation strategy for long-lived perennials, recalcitrant species and vegetatively propagated species. Their main limitation is that they take a great deal of space and are difficult to maintain and protect from natural disasters. They are susceptible to the spread of diseases and may suffer from neglect. Furthermore, out-breeders require controlled pollination for regeneration from seed. In many circumstances they are the only available option for the conservation of important germplasm. When displayed, the plants have an important educational value and can easily be accessed for research purposes.

In vitro Storage Methods

The storage of germplasm in laboratory conditions (in vitro) is specially suited for the long-term conservation of recalcitrant species and vegetatively propagated species. They can be stored at low temperature under slow growth conditions or cryopreserved in liquid nitrogen at -196ºC. Cryopreservation has so far been successful withonly a relatively few species but is a very promising development for long-term storage (WCMC, 1992). The main limitation of in vitro storage is the need for special equipment, techniques and trained staff. The cryopreservation of orthodox seeds represents an advantage over low-temperature storage at -20ºC, both economically and in terms of viability, as liquid nitrogen is a relatively inexpensive cryogen and the seeds retain the same viability as they had immediately before storage. Therefore regeneration costs are lowered and viability testing is reduced. However, more research is necessary to define the mechanisms of desiccation and chilling injury and to investigate methods of alleviating it (Eberhart, Roos & Towill, 1991).

Pollen Banks

Pollen preservation may be useful for base collections of species that do not produce orthodox seeds. It requires little space but some cytoplasmic genes would be lost. Like seeds, pollen can be divided into desiccation tolerant and intolerant. However, information about storage characteristics of pollen from wild species is fragmentary, existing mainly for some crop relatives and for medicinal and forest species (Eberhart, Roos & Towill, 1991).

DNA Banks

The creation of a network of DNA banks (DNA Bank-Net) to complement activities already being undertaken in ex situ conservation and more precisely germplasm collections can allow large quantities of genetic resources (genes, DNA) to be stored quickly and at low cost and could act as an insurance policy against rapid loss of the world's gene pool. It could be used in molecular phylogenetics and systematics of extinct taxa, and genes can be distributed via the polypeptide chain reaction (PCR) using primers supplied by the users (Adams et al., 1994). DNA samples have been mainly used for bio-prospecting and assessment of biodiversity studies. Its use in conservation is limited as whole plants cannot be reconstituted from DNA but the genetic material can be introduced to other genotypes for plant breeding and enhancement purposes. Its potential remains promising and should still be investigated as DNA is a very stable form in which genetic information can be stored (Ashton, 1988).


An essential part of any ex situ conservation method is the accurate documentation of the accessions as regards the origin and provenance of the collection along with relevant field data and subsequent information such as regeneration, viability testing, distribution and propagation (Given, 1987). Peeters and Williams (1984, in Ashton, 1988) estimated that 65% of the 2 million plant germplasm accessions held in genebanks worldwide lack basic data on the source and a greater proportion of those accessions lack characterisation and evaluation data vital for its subsequent use in plant breeding and enhancement. This has to be improved, because if no information is available on the resources conserved, chances are that they will never be requested and used in research and plant improvement programmes which should be the ultimate goal of any conservation strategy.

International Organisations Involved in Seed Banking/Genebanking

Today, as a result of the continuous efforts of botanic gardens for the last two hundred years in collecting plant material, and the great efforts on crop germplasm collection during the 1970s and the 1980s, there is a large number of genebanks and germplasm collections around the world. According to the FAO World Information and Early Warning System (WIEWS) database, approximately six million accessions are contained in over 1,300 genebanks around the world with 5.5 million accessions stored in regional or national genebanks. The number of seed stores has dramatically increased over the last 20 years going from 54 in the 1970s to more than 1,300 today, with nearly 400 with long-term and medium-term conditions. 60% of the total number of accessions are known to be stored in medium-term or long-term facilities, 8% in short-term facilities and 10% in field genebanks, in vitro and under cryopreservation. Clearly, seed storage is the predominant form of plant genetic resource conservation, accounting for about 90% of the total accessions held ex situ (FAO, 1996).

It is very difficult to give an estimate of the type of collections stored around the world as such information is known for only a third of the accessions in the WIEWS database. However, it has been estimated that 48% of all accessions are advanced cultivars or breeders' lines, while over a third are landraces or old cultivars and about 15% are wild or weedy plants or crop relatives (FAO, 1996).

Genebanks experience many limitations and constraints in efficiently conserving the material. Among these is the high number of recalcitrant species which cannot be stored in seed banks. These species therefore tend to be under-represented in germplasm collections. Furthermore many genebanks operate under very limited resources and under conditions of storage that often jeopardise the genetic integrity of the accessions conserved. For these reasons a large growing backlog of material is now due for regeneration. Incomplete data and limited access to information on conserved accessions can result in a significant amount of samples being over duplicated in different genebanks while at the same time a large number of other accessions not safely duplicated. Conserved material must be evaluated for the characteristics users want and most of this material still lacks legal protection under national legislation or international arrangements to ensure its long-term security.

Following on the pioneer work of the Russian N.I. Vavilov in the early 1900s in the conservation and utilisation of plant genetic resources, EUCARPIA (European Association for Research on Plant Breeding, established 1956) and FAO were among the first organisations to recognize the dangers of genetic erosion of the major crops. FAO with its Commission on Plant Genetic Resources aims to ensure the conservation, evaluation and utilisation of economically and socially important plant genetic resources. The International Plant Genetic Resources Institute (IPGRI), established in 1974, co-ordinates the efforts of the Consultative Group on International Agricultural Research (CGIAR) on the conservation of plant genetic resources. The CGIAR is a decentralised structure made up of 16 international agricultural research centres (IARC's) located in 12 developing and 3 developed countries and has a mission to promote sustainable agriculture for food security in developing countries. Expertise from population geneticists and organisations such as FAO and IPGRI are greatly contributing to the conservation of thousands of plant species. Of the 6 million accessions stored worldwide, some 600,000 are maintained with the CGIAR system (12 IARCs).

Significant advances have been made in conserving the plant genetic resources of the major crop plants of the world but much work remains. There are important gaps in the collections of minor crops and underutilized species worldwide. Groups of wild species with particular interests apart from crop relatives are medicinal plants, culinary herbs, ornamentals and landscape plants, those needed for habitat restoration and forestry species (Heywood, 1993). Such species are frequently lacking in other ex situ germplasm collections. Botanic gardens can fill an important gap in ex situ conservation and may even have a comparative advantage in conserving some vegetatively propagated plants and those with recalcitrant seeds, more particularly, trees and other slow-growing species. However, there is a need for botanic gardens to follow more rigorous methods of conservation and learn from the expertise of crop genebanks. The need for a comprehensive approach to ex situ conservation and the inclusion of botanic garden and arboreta in such programmes was emphasised by many countries in the sub-regional synthesis reports prepared for the International Technical Conference on Plant Genetic Resources that was held in Leipzig, Germany, in June 1996 (FAO, 1996).

International Survey of the Seed Banks/Genebanks in Botanic Gardens

There are more than 1,700 botanic gardens and institutions worldwide holding plant collections that serve both conservation and educational purposes. Many gardens have the mandate to preserve rare or threatened wild plants and make the plant material available for research. However, detailed information on their ex situ conservation methods such as seed banking was lacking. Therefore a review of the available information on botanic garden gene/seed banks, their collections and facilities, was undertaken from data held in the Botanic Gardens Conservation International (BGCI) databases and by analysing the data from the responses to a questionnaire sent to about 1,500 botanic gardens world-wide. Of these, 388 replies from gardens were received and analysed of which 152 reported having a seed bank/genebank. The questionnaire was divided into four main sections with enquiries about

  • the type and nature of accessions stored
  • the seed storage facilities and conditions
  • data processing and capacity to hold a database
  • the purpose of the collections
  • methods being used at that time by botanic gardens for the maintenance of germplasm collections, including seed banks, field genebanks and tissue culture collections, and to analyse the relative merits and limitations of each system.

This information can be used to make appropriate recommendations for their improvement and serve as a foundation to develop a strategy for seed banks/genebanks in botanic gardens as well as an international network of germplasm databases. A summary of the results can be found in Table 1.

Results of the Survey, 1994

The main conclusions:

  • Most of the seed banks surveyed are in warm temperate climates, which includes Europe.
  • Distribution is currently one of the main objectives of seed storage in botanic gardens with a strong emphasis on conservation in long-term seed banks.
  • 30.7% botanic gardens surveyed have some form of cool to low-temperature seed storage.
  • There is a great variation in the number of accessions representing each species. Long-term storage seed banks collect twice as much sample populations per species than medium and short-term seed banks (i.e. 36 species per 100 accessions in long-term seed banks as opposed to 75 species per 100 accessions in short-term seed banks).
  • At least 255,832 accessions of germplasm are stored by the 152 seed banks/genebanks surveyed and most of them hold species that are rare or endangered. There are 17,096 known accessions of germplasm held in field genebanks.
  • Long-term seed banks maintained by botanic gardens have approximately 77% of their collections of germplasm directly collected from the wild.
  • Medium-term seed banks have a higher proportion of crop plant species than long-term or short-term seed banks.
  • Most seed banks which have a collection from a specific geographical area, concentrate on the conservation of their regional flora.
  • Collections from a specific taxonomic group in long-term seed banks include families of major and minor crop plant species.
  • Few seed banks mention having a drying room and even less undertake moisture content tests.
  • Paper envelopes are mainly used by medium and short-term seed banks as packing material whereas laminated foil bags are mainly used by long-term seed banks.
  • 65.5% of seed banks have accessions intended to be stored for more than 15 years whereas only 27.5% of them store some of their accessions at temperatures below -15oC.
  • 28.8% of short-term seed banks have tissue culture facilities and 45.0% have an in situ reserve attached to their gardens.
  • 62.7% of seed banks/genebanks use a computer database system to store information generated from germplasm accessions.
  • 82.6% of long-term seed banks have a distribution policy but only 34.5% of the short-term seed banks, even though distribution is their main purpose.
  • Half of the seed banks/genebanks surveyed undertake seed germination testing.
  • Most botanic garden seed banks/genebanks feel that they would benefit from information exchange with other seedbanks/genebanks in botanic gardens.


A major concern arising from this survey is the drying rate of the accessions stored for medium and long-term storage and the type of containers used. The questionnaire failed to clearly assess the different methods used by the botanic gardens. However, it can be assumed that most gardens do not dry the seeds adequately to achieve long-term conservation. Additionally, very few keep their accessions in hermetically sealed containers. Organisations such as IPGRI have been investigating many of the major crop plant species and can provide information related to seed storage. This information can be used as a guide for the preservation of wild species which will require more experimentation mainly based on trials and errors.

Most of the seed banks surveyed maintain their collections under more than one storage condition, some are stored at low temperatures, some at cool temperature and some are stored at room temperature. This is either to safeguard the valuable germplasm or to simplify the work. Currently, medium-term seed banks might only be limited by technology from becoming long-term seed banks. There is clearly a need for more information on seed banking in botanic gardens. Furthermore, short-term seed banks will have to improve on the conservation value of their accessions in order to distribute seeds of a greater genetic variability.

An emphasis should be put on the conservation of useful plants such as wild crop relatives, primitive landraces and minor crops that would not be covered by the network of crop genetic resources. Rare and endangered plant genetic resources should be safely stored in a different place and crop genebanks would be adequate for the storage of duplicate collections. This should be the main focus for the development of a strategy for seedbanks/genebanks in botanic gardens closely linked with the crop genetic resources sector.

It is important to acknowledge that for conservation purposes, collections from well documented known wild sources are necessary and garden origin accessions should gradually be replaced. It is also important to increase the conservation value of seed accessions to maximise the use of facilities, technology and expertise. There should also be stricter quarantine control over the distribution of seeds through the Index Seminum system to avoid weed problems.

Seed banks/genebanks acquire and generate information at every stage of seed processing. They rely on a documentation system to store, maintain, process, analyse and exchange this information. This information is vital in seed bank management in order to set priorities, plan activities and manage resources. This can be the main limitation to establishing a network of plant germplasm accessions.

There is an urgent need for more guidelines on seed banking as well as sampling strategies with an emphasis on cost effective storage facilities for small scale genebanks as well as training of scientific and technical personnel. The gaps in existing collections should be identified in order to set collection priorities and research should be undertaken for non-orthodox species.

Further investigation of the known seed banks/genebanks should be undertaken and a list of their accessions available in the ITF format (International Transfer Format for Botanic Garden Plant Records). Conservationists and policy makers should use this survey to develop a global strategy for the conservation and utilisation of plant genetic resources.

Conclusion and Recommendations

A botanic garden which wishes to start a small seed bank /genebank would be advised to start with collecting germplasm that is very well documented from their living plant collection. This would allow them to experiment with a wide range of species and find suitable facilities and techniques for their particular needs. Once the set up is organised and functional, it would be advisable to collect accessions directly from the wild in order to distribute a wider genetic variability and to reduce the effect of domestication on the genetic make up of the accessions.

There is a need to generally improve the quality of seed banks in botanic gardens by improving the conservation value of accessions stored, the storage conditions, particularly the drying conditions and temperature of storage, and improve documentation systems and therefore their capacity to network with other plant genetic resources networks by adopting ITF and IPGRI documentation standards.

Botanic gardens should contribute to the implementation of the Convention on Biological Diversity (CBD) by sharing benefits and transfer of information generated by research. As nobody is self-sufficient in their need for plant genetic resources, all parties should be involved in sharing and exchanging resources and information. However, contracts should be agreed and signed between donors and receivers and benefits from products developed from the germplasm collected should be shared between the parties, in line with the provisions of the CBD. There is a need for clearly focused ownership, handling and distribution accession policies in botanic gardens.

Botanic gardens and the crop genetic resources network should collaborate in collecting threatened species and the range of diversity of a species, facilitating germplasm distribution and in research into genetic diversity such as eco-geographic studies, species mapping and the wild relatives of priority crops.

A network of databases of wild species should be developed to include well documented accessions from botanical institutions as well as other institutions such as the International Agriculture Research Centres of the CGIAR with plant genetic resources conservation activities coordinated by IPGRI. In order to create such a network, it is necessary to improve links between botanic gardens and major organisations involved in seed banking and promote exchange of information as well as technology.

Botanic gardens should also contribute to the establishment and running of regional genebanks in their area in order to avoid duplication of work and efforts. This would involved close links with the local communities and awareness of its needs.

It is of primary importance to create a network of seed banks for wild plant conservation. There is a need for guidelines for the management of small collections of wild species with limited resources. Botanic gardens have to improve their capacity to store the information generated from the seed banks on computer database systems in order to create international data exchange systems to allow an overview of botanic garden seed holdings to be gained.

BGCI, IPGRI and FAO should be involved in the establishment of a global network of plant genetic resources which would include the accessions of botanical institutions. This survey can be used to prepare a draft strategy for the development of an International Botanic Garden Seed Bank Network, including a list of long-term data requirements and a forum for the exchange of ideas and information and to help create new institutional links with the crop genetic resource sector.