Page 358

The Millennium Seed Bank at the Royal Botanic Gardens, Kew

Ghillean T. Prance
Roger D. Smith
Royal Botanic Gardens, Kew, Richmond, Surrey, TW9 3AB United Kingdom

Introduction

In response to the increasing rate of extinction of plant species (Ehrlich and Ehrlich 1981; Prance and Elias 1977), the Royal Botanic Gardens, Kew, decided in 1972 to begin seed-banking and research on seed conservation. Accordingly, the seed bank was set up at our second garden in the country at Wakehurst Place in Sussex, a safe location removed from a major urban area, not under the flight path of Heathrow Airport, and at an altitude of 200 m, well above sea level. Although we regard in situ conservation as the ideal, the world will certainly lose many species if we do not promote ex situ methods as well.

Seed banks are one of the most effective and economical means of conserving plant species where habitats are under threat (Miller and others 1995). In the last 2 decades, the current Kew Seed Bank has undertaken collaborative collecting expeditions in over 20 countries, and collecting activity has increased substantially over the last few years. These efforts have made the Kew Seed Bank the largest and most diverse bank that is devoted to wild plants and is run according to internationally approved standards. However, because financial resources are sparse, the bank collection still represents less than 2% of the world's flowering-plant flora. Viewed against the background of a rapidly increasing loss of biodiversity, that prompted us to investigate the possibility of increasing even more the rate of seed conservation and seed-banking undertaken by Kew.

In 1994, aided by a consultant, Sir Jeffery Bowman, we carried out a detailed study of the worldwide situation. Our survey indicated that there was very little

Page 359

coverage of noncrop plants in seed banks and minimal research into optimal collection, processing, and storage procedures for such species. That finding was supported by a recent review of the state of the world's plant genetic resources for food and agriculture by the Food and Agriculture Organization (FAO 1996), which concluded that there was a clear need to strengthen capacities for ex situ conservation cost-effectively and that the necessary increase in seed-banking activity and supporting research would require national, subregional, regional, and international collaboration. The report also confirmed the heavy emphasis (94%) on crop plants among the 6 million seed accessions held worldwide, the minimal coverage of truly wild species, and the slight coverage of forest, forage, ornamental, aromatic, and medicinal species and underused crops. Moreover, the FAO report highlighted the fact that only 13% of the 6 million accessions were held in secure long-term facilities—that is, where seed was stored according to internationally approved standards of temperature and moisture content, where the power supply was reliable, and where procedures for safe duplication and regeneration were in place. The current Kew Seed Bank, which was mentioned specifically in the FAO report, meets all those criteria for a secure, long-term seed bank.

On the basis of such information and with impetus added by the UK ratification of the Convention on Biological Diversity (UNEP 1992a), Kew concluded that a substantial increase in efforts to collect, conserve, and research seeds of wild species was vital. Moreover, Kew feels that it is uniquely placed to play a leading role in this process, not only because of its existing collections, its expertise in seed conservation, and its location within a geologically and politically stable country, but also because of its well-established network of collaborators and its horticultural and taxonomic expertise, which has earned it an international reputation as a center of excellence for botanical research. That expertise and a belief in collaboration will be vital to ensuring the international cooperation needed to tackle a problem of this scale.

The opportunity to achieve the great increase in Kew's seed-conservation activities was provided by the Millennium Commission, one of the distributors of national lottery proceeds in the UK, which was set up for partial funding of projects to celebrate the new millennium. In December 1995, the commission awarded Kew's Millennium Seed Bank (MSB) project a grant that would eventually total up to £30 million, which is just over one-third of the total cost of the project. With the help of the Kew Foundation, we have raised over £16 million in counterpart funding, including a grant of £9.2 million from the Wellcome Trust and a sponsorship of £2.5 million from Orange, a UK communication company. The MSB will continue to focus on wild species rather than crop species, many of which have their own seed banks or germplasm collections.

The MSB project has been presented to and discussed with representatives of national and international organizations involved in plant genetic-resources conservation, including FAO, the Consultative Group for International Agricultural Research, the International Plant Genetic Resources Institute, Botanic Gardens Conservation International, the World Conservation Union, the United Nations Environment Program (including the Secretariat to the Convention on Biological

Page 360

Diversity), the United Nations Development Program, the Global Environment Facility, and the World Bank. In addition, relevant UK government departments (Department for International Development formerly ODA, Department of the Environment, and Ministry of Agriculture, Fisheries and Food) and conservation bodies have been consulted, and the proposal has been presented at several scientific conferences. All such meetings have confirmed not only that a largescale seed-conservation project is necessary and would not duplicate any existing activity but also, inasmuch as Kew is a world leader in seed-banking for wild plants, that it is ideally placed to be the focus for such a major conservation effort.

The Aims of the Millennium Seed Bank

The MSB project will establish an international center of excellence for seed conservation at the Royal Botanic Gardens, Kew, Wakehurst Place. The project has six main aims:

• to collect and conserve seeds of most of the UK spermatophyte flora (seed-bearing plants) and a further 10% of the world's spermatophyte flora, principally from the drylands;

• to encourage plant conservation throughout the world by facilitating access to and transfer of seed-conservation technology;

• to carry out research to improve all aspects of seed conservation;

• to make seeds available for species reintroduction into the wild, for academic research, and for screening for potential new uses of plants;

• to develop the public's interest in the need for plant conservation; and

• to provide a world-class building as the focus for this activity.

The UK Seed Conservation Program

For Kew to function actively in seed conservation overseas, it is important that it make an input into plant conservation within the UK, where genetic erosion and endangerment are also high (Anon 1994; Wynne and others 1995). Common species will be included to supply material off-season or abroad, to add seed-biology information, to compare with in situ populations through time, and to guard against changing fortunes resulting from climate change (see Jackson and others 1990). No country holds a near-complete representation of its spermatophyte flora. Kew aims to enable the UK to be the first such country and hopes that the example will stimulate other countries to follow suit.

Our initial objective is to have conserved within the MSB, by the year 2000, seed from at least one population sample of every native UK plant species that produces bankable seed.

Stace (1991 and pers. comm.) has indicated that the native flora of the British Isles consists of some 1,571 species of vascular plants, of which 1,442 are spermatophytes native to the UK. The remainder are ferns or plants that occur only in Eire. Those figures do not include the microspecies of the apomictic

Page 361

(reproducing without meiosis or formation of gametes) genera Rubus, Hieracium, and Taraxacum, which have been grouped together into the more distinct aggregates numbering 13, 11, and 9 respectively. The Kew Seed Bank holds 552 species (plus some microspecies), so 890 need to be collected; 361 of these are restricted or rare. Within the target for collection, we estimate that six species produce seeds that cannot be banked (the recalcitrant species), and 11 rarely or never produce seed. That leaves 873, of which 163 (including 77 aquatics and 51 orchids) produce seed that will need research work before we can ascertain the likely success of banking.

Seeds for the MSB will be collected in collaboration with many individuals and conservation organizations throughout the UK, including the statutory bodies—English Nature, Scottish National Heritage, the Countryside Council for Wales, and the Department of the Environment (Northern Ireland)—and those like the Botanical Society of the British Isles and the Wildlife Trusts. A member of our collecting staff will work full-time on the project for the next 2½ years, but many of the remaining collections will be made by partner organizations; their work will be coordinated by the MSB. Training sessions for volunteer collectors from within the partner organizations are under way.

Where appropriate, difficult species will be collected through a contract arrangement with specialist organizations, and the more common species will be collected by voluntary groups, such as the Wildlife Trusts, and will attract the payment of an honorarium. There will thus be high public involvement. If the seed-production seasons are abnormal, it might be necessary to get the volunteers to retry some species the next year. It is envisaged that the less than complete genetic representation (especially of inbreeding species) that results from stopping at the initial objective of one population sample per species will be gradually improved by donation to the collection and by further collecting later in the project.

Target-species lists have been provided to the Wildlife Trusts and several other organizations to induce offers of collection. These lists have been produced from an extensive Excel database, now substantially developed, that lists all the UK native spermatophytes by their scientific and common names broadly where they occur, the status of existing collections, their rarity, and any special problems related to them (for example, that they are aquatic). It is proposed that other useful information, such as flowering or seeding date, be added to the database in due course. By May 1999, seeds from 71% of the species that are native to the UK have been collected and booked.

The Arid-Land Seed-Conservation Program

Since the early 1980s, the focus of the Kew Seed Bank has been on tropical drylands. Such lands, which are experiencing habitat loss because of desertification, particularly in Africa (UNEP 1994), have been identified as the ecosystem for which ex situ conservation is most appropriate, compared with the tropical rain forests (the other ecosystem undergoing extensive damage). The origin of habitat loss in the drylands—drought and the factors that exacerbate it, such as overgrazing (Binns 1985)—is less open to substantial manipulation with local political

Page 362

or economic tools. Consequently, other actions are necessary to underwrite the survival in situ of biological diversity in the tropical drylands.

The drylands cover one-third of Earth's land surface, including many of the world's poorest countries, and support almost one-fifth of its human population (UNEP 1992b). Rural people rely on plants in almost every aspect of their lives, and Kew's Survey of Economic Plants for Arid and Semi-Arid Lands (SEPASAL) database lists over 6,000 such plants with uses as varied as land stabilization, hedging, nitrogen fixation, contraceptives, dyes, and cooking utensils (Davis and others 1996). Products derived from plants in the drylands are also important to people in developed countries, for example, the pharmaceuticals sennoside A and B, atropine, and ephedrine and such industrial products as gums, resins, waxes, and oils (Goodin and Northington 1985). There is great scope for many more dryland plants and their products to be developed for human welfare, including those with unique morphological, physiological, and chemical adaptations induced by the particular environmental stresses of arid lands. Some of these adaptations, such as salt tolerance and the C4 and CAM mechanisms of photosynthesis, might be valuable sources of material for plant-breeding. Other characteristics that confer tolerance to drought, predation, and disease are also likely to be present in dryland plant material. In addition, many arid-land species have evolved elaborate chemical defenses that make them important potential sources of insecticides.

Some practical benefits of seed-banking follow the choice of the drylands as a target for seed collection: species are often found within discrete populations; populations often exhibit defined flowering and fruiting periods in response to climatic conditions; and vegetation is usually low (often less than 5 m high) and relatively open, allowing convenient access for seed collection. Furthermore, although the seed-storage physiology of only 2% of the world's flora has been studied, it is thought that the majority of higher plant species from drylands will exhibit “orthodox” seed-storage behavior (retain their viability after drying) and therefore be suitable for long-term conservation in seed banks.

Prospective countries have been identified for partnership on the basis of several factors: existing successful collaboration; ease of access; extent of arid, semiarid, and dry subhumid land (Goodin and Northington 1985); number of endemic plant species (WCMC 1992); and the floristic regions in which they occur (Taktahjan 1986). On that basis, 18 countries have been identified as having high priority for collaboration; of these, eight (Australia, Brazil, Kenya, Madagascar, Mexico, Morocco, South Africa, and the United States) contain the most diverse dryland floras (table 1). In addition, we will accept donations of seed from nondryland countries (or from the wetter regions of the high-priority countries), provided that collections have been made in accordance with national and international regulations.

The world's spermatophyte flora is estimated to number 242,000 (Mabberley 1990); 10% is therefore 24,200 species. Within the collaborating countries, success in collecting and conserving seeds of 10% of the world's plant species will depend heavily on input from the countries' conservation organizations themselves; the overriding consideration in any targeting of taxa for collection will be

Page 363

our partners' priorities for conservation, in line with the Convention on Biological Diversity.

Nevertheless, it was recognized early in the project's development that some focusing of collecting activity is important. For the last few years, the collectors for the current Kew Seed Bank have given special interest to 30 plant families selected on the basis of an analysis of species represented in Kew's SEPASAL database (table 2).

This list of target families is now being reviewed and revised to include coverage of globally threatened species and endemics, with input from the World Conservation Monitoring Center and partner institutes in collaborating countries, such as the National Museums of Kenya, and adequate representation of “higherlevel” (order and above) taxonomic diversity as a surrogate for character and evolutionary diversity (see, for example, Williams and others 1994). It is also intended to give due weight to various ecological and functional considerations, such as appropriate representation of “keystone species” in particular plant communities, as more becomes known about such species.

Obviously, some flexibility will be essential to take account of changing circumstances. We expect a continuous process of review and refinement of any lists of desiderata as the project develops.

The arid-land collecting program will build on existing links between the current Kew Seed Bank and institutes in many dryland countries established during collaborative collecting expeditions in over 20 countries in the last 2 decades. The main planning phase for the overseas conservation program will be in 1996–1999; the main collecting phase will be in 1999–2009. Initial efforts will focus on seeking collaboration with the high-priority countries, including the United States, so that seed-collecting in these countries as part of the project could start by the

Page 364

year 2000. For the remaining dryland countries, contacts with those with which we are collaborating or have collaborated in the past will be maintained or renewed, and contacts will be established in the others during the next 2 years with a view to securing partnership for later in the project period.

Page 365

The needs of collaborating countries will differ according to their current levels of expertise and their national priorities for biodiversity conservation, so the MSB project will aim to provide a comprehensive seed-conservation service to meet varied requirements. The key services offered to collaborators are training and technology transfer, long-term storage of seeds until facilities exist in their own countries, benefit-sharing from the use of seeds, and access to our research expertise and data.

All seed-collecting, storage, and distribution will be carried out under bilateral agreements that cover profit-sharing as a result of intellectual property rights. The agreements have been developed as part of Kew's institutional program to ensure full compliance with the Convention on Biological Diversity. We are also in consultation with Michael Gollin (attorney at Spencer & Frank, Washington, DC, who has particular experience in contracts related to pharmaceutical screening) and Neil Hamilton (director of the Agricultural Law Center, Drake University Law School, University of Iowa).

The project will increase the number of collectors from the current two to around 28 in the year 2000. Five collectors and a coordinator will be based at the MSB; the remainder will be based overseas in partner countries. It is hoped that most of the overseas-based collectors will be recruited from the collaborating countries to collect their national floras and will be funded by international donor agencies, such as the Global Environment Facility and the European Union.

The key aim is to sample the breadth of dryland plant diversity, concentrating on a wide range of species. Interspecific variation is often the initial screen for potential use. Although only one population sample per species will be collected as a key objective, substantial genetic variation is likely to be present in collections; most species will be outbreeders (see Richards 1986), and intrapopulation variation will be greater than with inbreeders (von Bothmer and Seberg 1995). The samples collected can be used to research species botany, including breeding system, and so more carefully tailor later sampling strategy. To sample genetic diversity within a species more fully, more populations would need to be collected (Brown and Marshall 1995), but that is more appropriate after successful initial trials or research. The sampling strategy for each population will be that practiced for over 20 years by the Kew Seed Bank and is similar to that recently reiterated by Brown and Marshall (1995); key factors will be to sample randomly and evenly within a population and from sufficient individuals (at least 50 where the size of the population permits). Where populations are very small (20 or fewer), collections from individual plants will be kept separate—something that is impractical for more individuals. Collectors will visit most populations once during the seeding period, so it is proposed that no more than 20% of the seed available on the harvest date should be taken from annuals, biennials, and short-lived perennials. Thus, the survival of the parent plant population should not be threatened.

In achieving the 24,000-species target, allowances have been made for duplicate and poor-quality collections. In addition, it is envisaged that 10% of the target will consist of unsolicited samples, some from within the target area and many from outside.

Page 366

Voucher specimens, representing the population sampled, are always collected. At least one of these specimens is deposited in the national herbarium of the host country. The specimen returned to the MSB will be identified by reference to Kew's comprehensive herbarium collection and associated bibliography. Data recorded in the field will be as objective as possible. All populations will be located with Global Positioning Systems, either in the partner countries or in the MSB. Such systems will be used, where possible, to guide collectors to likely diversity hot spots and, in some circumstances, to predict where a particular species might be found (see Guarino 1995). They might even be able to guide collectors to areas where the greatest genetic diversity would be expected within a species range. For instance, Nevo and Beiles (1989) hypothesize that species ranging across mesic and xeric environments display the greatest levels of genetic diversity in hot deserts, where rainfall and climate unpredictability are highest.

All collections destined for the MSB will be returned from the field as quickly as possible, thereby minimizing the loss of initial seed viability, which can strongly influence longevity (see Smith 1995). Samples thereafter will be processed as they are in the current Kew Seed Bank (see Prendergast and others 1992). The procedures are adapted from those used by seed banks that store crop germplasm (see Ellis and others 1985). They involve accessioning, drying, cleaning, x-ray examination, counting, packaging, freezing, and testing of germination. The main differences between processing of wild and crop germplasm are related to the handling of “empty” or insect-damaged seed (Linington and others 1995), seed dormancy (Linington and others 1996), and identification of seed storage behavior by testing germination after drying and freezing (Smith and Linington 1996). Storage of seed will be mainly at -20°C in a variety of storage containers. Subsamples will be rechecked for germination initially every 10 years, as in the current Kew Seed Bank, but retest intervals are expected to be modified for different collections in the light of results from the seed-research program of the MSB project. It is expected that 22 processing staff will be based in the collaborating countries and a further 22 in the UK.

Seed samples collected will be shared equally between partner countries and the MSB and deposited in facilities in the country of origin, if available. The MSB will act as a backup, providing a duplicate store for an agreed proportion of the seeds. For countries where local facilities are not available, the MSB will store independently both partners' shares of each collection and provide advice and assistance on establishing a bank in the country of origin. The MSB will, in turn, back up some of its collections at the Scottish Agricultural Sciences Agency, East Craigs, Scotland, to achieve a double indemnity against loss.

Seed-Banking in the United States.

The United States contains a considerable area of arid lands, and three collaborative expeditions have already taken place in the country. A meeting was held in November 1996 between one of us (R.S.) and Peggy Olwell, chairperson of the Native Plant Conservation Committee, a partnership of nine federal agencies and 54 nonfederal cooperators, to discuss the MSB project. Considerable support for

Page 367

the project was shown, and an invitation was extended to attend its bimonthly meetings and become a nonfederal member. Contact about the MSB has also been established with the Center for Plant Conservation (CPC), a nonfederal member of the above committee and an entity with which we have previously collaborated on technical matters in wild-species seed-banking. Michael Bennett, keeper of the Jodrell Laboratory, gave a presentation on the MSB during his visit to the Missouri Botanical Garden (home of the CPC) in May 1997. Again, an interest in collaborating was expressed. Similar enthusiasm for involvement with the MSB has been shown by representatives of the Boyce Thompson Southern Arboretum, the Desert Botanical Garden, and the University of Arizona Desert Legume Program.

A meeting with the staff of the US Department of Agriculture's National Seed Storage Laboratory at Fort Collins to discuss collaboration with the MSB took place in August 1997 to coincide with a conference on plant genetic resources.

Research, Information Flow, Technology Transfer, and Training

For the MSB project to succeed, it must not be merely a museum of seeds. It will also be accompanied by considerable research on seed-collecting, processing, and storage and by extensive training and, where needed, technology transfer to our collaborators.

The integrated seed-banking and seed-conservation research program of the MSB offers a unique opportunity to increase knowledge on the seed biology of a considerable amount of dryland biodiversity. The value of this information will be fully realized if it is made readily available to potential end users, and this will be facilitated through technology transfer and the provision of advice and training. Thus, this part of the MSB project will have the following main objectives:

• To generate detailed primary datasets on the seed storage and germination of about 1,500 species through research.

• To construct a seed-information database on about 20,000 species using inhouse and public-domain information.

• To ensure benefit-sharing through information flow, technology transfer, and formal training.

Research

The Seed Conservation Section has been involved in research in the conservation of seeds of wild (nondomesticated) species for over 25 years. The research has already established that wild species differ from domesticated species in their seed-storage and germination behavior, but their conservation and use as seed are generally practicable.

The seed-storage behavior of only about 7,000 species has been investigated to any degree of certainty; and of the remainder of the world's spermatophyte flora, an estimated 37% of species occur in families where less than 1% of those species have had their behavior investigated (derived from Hong and others 1996).

Page 368

Moreover, the dryland floras are among the most poorly known botanically of all the biomes (Frodin 1984), and details of seed characteristics of dryland species are similarly restricted to about 300 species (Gutterman 1993). Thus, with the key aim of sampling the breadth of dryland plant diversity, the MSB project will inevitably be dealing primarily with species new to seed-conservation science. Two main problems are envisaged. First, about 4% of species handled are not expected to be readily suited to conventional storage protocols, and modifications might be required of all phases of the banking activity (collecting, processing, and storage) to ensure their conservation. Second, a smaller proportion of species are likely to require detailed investigation of their germination requirements so that their genetic potential can be readily released. In addition, research on predicting seed longevity of bank collections will be required as a management tool for setting seed-viability retest intervals to improve the balance of seed consumption during the monitoring of viability and the need to maintain seed stocks. Overall, it is envisaged that the total number of species requiring research on germination and storage will be about 150 per year. There will be 21 research staff and space for 15 visiting researchers.

Research to improve collecting. The collecting phase of species conservation offers the first opportunity to identify potential seed-storage problems and apply modified handling procedures to ensure that seeds retain maximal quality before seed-banking. Thus, the current research program in this regard will be expanded and focus on improving the field diagnosis of seed-storage behavior and maximizing the harvest quality of the collections.

On the basis of information generated automatically during the processing of seed for storage in the MSB and other databases worldwide as they become available, a relational database of seed information will be constructed for over 20,000 species. It will be compatible with other databases within Kew (such as SEPASAL) and outside Kew and be used to develop a field diagnostic algorithm for potential routine seed conservation.

The vast majority (86%) of desiccation-tolerant seeds collected should remain viable for at least 200 years (Hong and others 1996) under international standard bank conditions. Historical data (see Bewley and Black 1994) and retest data from the Kew Seed Bank support that contention. However, although all collections made will be moved from the field to Wakehurst Place as quickly as possible, at least two aspects of their physiology might change before their arrival: desiccation tolerance and potential longevity (Hay and Probert 1995; Smith 1995). Developing methods to minimize and control such changes is of paramount importance if long-term storage is to be guaranteed for all collections.

Research to improve processing. The rapid and reliable distinction between the two extremes of seed-storage behavior (desiccation-tolerant and -intolerant) and subjecting of collections to appropriate processing are ever more urgent as the systematic ex situ conservation of species progresses.

Most bank collections undergo some field drying as part of the natural process of maturation of the parent plant. And it is often necessary to clean and partially

Page 369

dry some fleshy fruits for logistical reasons before dispatch and to reduce the opportunity for fungal infestation of the seed lot during transit to the bank. However, it has become clear recently that such postharvest practices have potentially large effects on the long-term maintenance of seed quality. For example, studies on crop seed have indicated that even a small alteration in moisture content could switch the physiological mode of the seed into or out of self-repair and hence affect seed quality. Also, the method of seed dehydration can have a profound bearing on the extent of desiccation tolerance.

To complement the first-step diagnosis of behavior in the field, more-detailed laboratory-based investigations are needed at a more mechanistic level to understand the process of desiccation intolerance. It would be an important advance, not only to seed conservation but also to seed science in general, if a universal set of markers of desiccation tolerance could be identified. Such a development would allow the screening of seed lots that had been identified by the field diagnostic as being of highest banking uncertainty to undergo rapid biochemical diagnosis.

Although studies, principally at the Kew Seed Bank, have resulted in the development of germination algorithms for many families, it is still estimated that a substantial number of previously untested species in a broad range of families will require further research to allow efficient germination. Families that do not respond to normal germination algorithms, such as Compositae (Linington and others 1996), and families for which no germination information exists require particular attention. This element of the research program will provide a unique opportunity to make detailed studies of dryland-species regeneration strategies and to model population responses to environmental cues, thereby substantially increasing our knowledge of dryland-seed biology.

Thus, our research objectives are to improve seed-drying methods, to continue the search for a biochemical diagnostic procedure for desiccation tolerance, and to develop effective germination-test regimens for dryland species further.

Research to improve storage. A large amount of seed is needed to quantify the longevity response at a single temperature, so there is also a need to provide a fundamental understanding of the mechanism of viability loss to develop a rapid and efficient system of diagnosing the storage potential of collections from a small quantity of seed.

Earlier sections of this review have clearly identified the need to quantify further how seed longevity in a species can be affected at all stages of the conservation process (collecting, processing, and storage). Of particular importance is the recent suggestion that the optimal storage conditions for orthodox seeds can differ (Vertucci and others 1994). Thus, long-term experiments, some of which should be at seed-bank temperatures, are required to establish whether there really are long-term implications of this suggestion. Our review of orthodox seeds (Hong and others 1996) has revealed that the potential longevity in storage under seed-bank conditions is known to vary considerably; predicted longevities vary over a factor of 200 in the 52 species representing 23 families for which seed-storage responses have been quantified sufficiently to allow comparison under identical conditions.

Page 370

Further quantification of the variation in the rates of viability loss is needed so that appropriate retest intervals can be set and unnecessary depletion of the collections avoided. Moreover, it is appropriate to consider the cause of the intrinsic differences in dry-seed longevity. Increasing evidence of nonorthodox seed-storage behavior across species of many families demands that nonconventional storage environments be considered for the storage of some species. It is predicted that 163 species in the native UK spermatophyte flora will be difficult to collect or conserve and will therefore require researching. A brief survey of dryland species suggests that a substantial number will possess nonorthodox seeds as a result of either low desiccation tolerance or sensitivity to seed-bank temperatures. Improved short-term storage protocols are required for whole seeds to allow highviability seed to be available as the starting material for long-term conservation. In addition, long-term storage of nonorthodox seeds under nonconventional temperatures for seed-banking, including cryopreservation, is suggested.

Seed-Information Database

The computerization of records for the seed-bank collections was started in 1981 and now includes information on nearly 11,000 accessions. Information recorded includes passport and management data. Passport data include date of collection, name and affiliation of collector (Kew or other), geographic location, type of material, number of individuals and proportion of population sampled, voucher and taxonomy, and distribution policy. For regenerated seed stocks, the following details are recorded: parent plant plus sibling data, generation, where grown, isolation conditions if any, number of seeds sown and number of plants harvested, date of harvest, voucher, and distribution policy. Management data cover x-ray record, bank location, original and current seed number, storage temperature, number and type of container, original and retest germination results and conditions, verification details and taxonomy, location of duplicate collections, interval of retest, and distribution date.

Some of the seed-bank database and summarized research datasets will be used to develop a relational seed-information database that should cover more than 20,000 species by the year 2010. Information will probably include the physical and chemical characteristics of the seed and optimal collecting, storage, and germination details. The database design is expected to ensure a high level of connectivity to other databases in and outside Kew, thus maximizing the potential use of the database as a management tool for our collections and as a means of providing advice on seed conservation to our collaborators and the public, for example, throughout data outlets in the public interpretation area (winter garden) of the MSB building.

Information Flow, Technology Transfer, and Training

Since 1992, the Seed Conservation Section has published over 90 scientific articles. In addition, the list of seeds that is used to publicize the material available for use and summarize the group's activities has been published biennially. More than 280 scientific visitors to our facilities have been accommodated since 1992;

Page 371

during the same period, the research group welcomed 12 foreign visiting scientists at undergraduate to postdoctoral levels. Moreover, collaborative projects were initiated with 20 institutes in the UK and abroad, including five of the eight countries with the highest-priority for collaboration with the MSB. The group has considerable experience in organizing conferences, running training courses (over 280 students have attended our formal training courses in the last 5 years), and supervising PhD and MSc student projects.

It is envisaged that each year up to 57 trainees or researchers from collaborating countries will visit the MSB for at least a month, and there will be shorter-term visitors. The new MSB building will include accommodation for up to 28 visitors from collaborating countries at any time. That will facilitate training and technology transfer, which we see as being achieved in a number of ways: advisory visits by MSB staff, for example, to help develop seed-storage facilities; opportunities for scientists to come to the MSB to gain practical experience in specialized seed-conservation techniques, such as the identification of high-quality seeds and long-term seed-storage techniques; and opportunities to attend formal training courses. Training will be available at all levels of expertise, from technician to postdoctoral, and for various periods, from 1 month to several years.

In addition to the data produced from the routine processing of and research on seeds, the MSB will provide collaborators with general information service on many aspects of seed conservation. Visiting scientists will have opportunities to access Kew's vast resources, including the herbarium and library, and, by arrangement, other parts of Kew, such as the Jodrell Laboratory and the Center for Economic Botany. In addition to existing databases, such as SEPASAL, further databases, such as the seed-information database, will be developed throughout the project for use by collaborators. Collaborators will receive updates of important developments in seed conservation, including details of the latest key publications.

Conclusion

The MSB is one of the most ambitious projects ever undertaken by the Royal Botanic Gardens, Kew. However, the biodiversity crisis that the world is facing calls for such large-scale remedies to avoid disaster. We have been encouraged by the response to the MSB project both by the public and by many sources of funding as expressed in the fact that within the brief period of 2 years of planning we have been able to obtain £45 million ($73 million) for the project to add to Kew's own commitment of about £8 million ($13 million). That seeds will be stored in both the MSB at Kew and seed banks of many collaborating countries must not detract from the need to maximize the efforts of in situ conservation, which allows species to continue to interact with their environment and allows the process of evolution to continue.

Page 372

Acknowledgments

We thank Gillian Wechsberg for compiling much of the information presented here. We are also grateful for the help of Simon Linington, John Dickie, Hugh Pritchard, and Robin Probert.

References

Anon. 1994. Biodiversity. The UK action plan. London: HMSO.

Bewley JD, Black M. 1994. Seeds: physiology of development and germination. New York: Plenum Press.

Binns T, editor. 1995. People and environment in Africa. Chicester, UK: John Wiley & Sons Ltd.

Brown AHD, Marshall DR. 1995. A basic sampling strategy: theory and practice. In: Guarino L, Ramantha Rao V, Reid R, editors. Collecting Plant Genetic Diversity, Technical Guidelines. CAB International.

Davis SD, Sinclair NJ, Cook FEM. 1996. The work of Kew's Center for economic botany and the survey of economic plants for arid and semi-arid lands (SEPASAL). In: West NE, editor. Rangelands in a Sustainable Biosphere. Proceedings of the Fifth International Rangeland Congress 1:111–2.

Ellis RH, Hong TD, Roberts EH. 1985. Handbook of Seed Technology for Genebanks, Vol. 1. Principles and Methodology. Rome: International Board for Plant Genetic Resources.

Ehrlich PR, Ehrlich AH. 1981. Extinction: the causes and consequences of the disappearance of species. New York: Random House.

FAO. 1996. The state of the world's plant genetic resources for food and agriculture. Rome: Food and Agriculture Organization of the United Nations.

FAO/IPGRI. 1994. Genebank standards. Rome: Food and Agriculture Organization of the United Nations, Rome and International Board for Plant Genetic Resources.

Frodin DG. 1984. Guide to standard floras of the world. Cambridge, UK: Cambridge University Press.

Goodin JR, Northington DK, editors. 1985. Plant resources of arid and semi-arid lands: a global perspective. London: Academic Press.

Guarino L. 1995. Geographic information systems and remote sensing for plant germplasm collectors. In: Guarino L, Ramantha Rao R, Reid R, editors. Collecting Plant Genetic Diversity, Technical Guidelines. p 316–28.

Gutterman Y. 1993. Seed germination in desert plants. Berlin: Springer Verlag.

Hay FR, Probert RJ. 1995. Seed maturity and the effects of different drying conditions on desiccation tolerance and seed longevity in Foxglove (Digitalis purpurea L.). Annals of Botany 76:639–47.

Hong TD, Linington S, Ellis RH. 1996. Seed storage behavior: a compendium. Rome: International Plant Genetic Resources Institute.

Jackson M, Ford-Lloyd BV, Parry ML, editors. 1990. Climate change and plant genetic resources. London: Belhaven Press.

Leprince O, Hendry GAF, McKersie BD. 1993. Seed Science Research 3:231–46.

Linington S, Terry J, Parsons J. 1995. X-ray analysis of empty and insect-damaged seeds in an ex situ wild species collection. IPGRI / FAO Plant Genetic Resources Newsletter 102:18–25.

Linington S, Mkhohta D, Pritchard HW, Terry J. 1996. A provisional germination testing scheme for seed of the Compositae. In: Hind DJN, editor. 1994. Proceedings of the International Compositae Conference, Kew. vol.2. Royal Botanic Gardens, Kew.

Mabberly DJ. 1990. The Plant-Book. Cambridge: Cambridge University Press.

Miller K, Allegretti MH, Johnson N, Jonsson B. 1995. Measures for conservation of biodiversity and sustainable use of its components. In: Hewwood, VH, Watson RT, editors. Cambridge: Global Biodiversity Assessment. Cambridge University Press.

Nevo E, Beiles A. 1989. Genetic diversity in the desert: patterns and testable hypotheses. Journal of Arid Environments 17:241–4.

Ponquett RT, Smith MT, Ross G. 1992. Lipid autoxidation and seed ageing: putative relationships between seed longevity and lipid stability. Seed Science Research 2:51–5.

Prendergast HDV, Linington S, Smith RD. 1992. The Kew Seed Bank and the collection, storage and utilization of arid and semi-arid zone grasses. In: Chapman GP, editor. Desertified Grasslands, their Biology and Management. London: Academic Press.

Page 373

Prance GT, Elias TS, editors. 1977. Extinction is forever. The New York Botanical Garden. p 437.

Richards AJ. 1986. Plant breeding systems. London: Unwin Hyman.

Smith RD. 1995. Collecting and handling seeds in the field. In: Guarino L, Ramantha Rao V, Reid R, editors. Collecting Plant Genetic Diversity, Technical Guidelines CAB International.

Smith RD, Linington, SH. 1996. Practical management of the Kew Seed Bank for the conservation of arid land and UK wild species. In: Proceedings of the Workshop on the Conservation of Wild Relatives of European Cultivated Plants. Council of Europe.

Stace C. 1991. New Flora of the British Isles. Cambridge: Cambridge University Press.

Taktahjan A. 1986. Floristic regions of the World (Translated: Cronquist A, editor). Berkeley, CA: California University Press.

UNEP. 1992a. Convention on Biological Diversity. United Nations Environment Program.

UNEP. 1992b. World Atlas of Desertification. London: Edward Arnold.

UNEP. 1994. United Nations Convention to Combat Desertification in those Countries Experiencing Drought and/or Desertification, Particularly in Africa. United Nations Environment Program.

Vertucci CW, Roos EE, Crane J. 1994. Theoretical basis of protocols for seed storage III. Optimum moisture contents for pea seeds stored at different temperatures. Annals of Botany 74:531–40.

Von Bothmer R, Seberg O. 1995. Strategies for the collecting of wild species. In: Guarino L, Ramantha Rao V, Reid R, editors. Collecting Plant Genetic Diversity, Technical Guidelines. CAB International.

Williams PH, Gaston KJ, Humphries CJ. 1994. Do conservationists and molecular biologists value differences between organisms in the same way? Biodiversity Letters 2, 67–8.

Williams RJ, Leopold AC. 1989. The glassy state in corn embryos. Plant Physiology 89, 977–81.

World Conservation Monitoring Center (WCMC). 1992. Global Biodiversity: Status of the Earth's Living Resources. London: Chapman & Hall.

Wynne G, Avery M, Campbell L, Gubbay S, Hawkswell S, Juniper T, King M, Newberry P, Smart J, Steel C, Stones T, Stubbs A, Taylor J, Tydeman C, Wynde R. 1995. Biodiversity Challenge. RSPB, Sandy.