around the world, and less formal collaborations
with 123 institutions in 54 countries. The Millennium
Seed Bank (MSB) currently holds accessions of more
than 28 000 species, including more than 10 379 ac-
cessions of 3 3 18 species with known food use.
The MSBP is working to overcome constraints to the
conservation and use of plants important to local
livelihoods. Germination tests have been carried out
on 3 028 taxa with food uses; 2 102 of these have >
75 percent germination, the current minimum MSB
standard for storage. Germination protocols are
made available via the Seed Information Database
(Royal Botanic Gardens Kew, 2008). Kew’s “Difficult”
Seeds project worked with African gene banks to
identify 220 species, most of them food plants, with
inherent seed storage problems, seed dormancy is-
sues, or poor viability due to inadequate handling and
storage. Training workshops included a two-day
mini-workshop for local farmers and community
representatives, with the aim of supporting and fa-
cilitating gene banks to engage with farmers. Es-
sential seed biology information for 160 “difficult”
species, together with training materials, is available
via Kew’s web pages (Royal Botanic Gardens Kew,
2 010).
MSBP partners are also working with local commu-
nities to document, collect, conserve and propagate
the genetic diversity of useful wild plants. The MGU-
Useful Plants Project works with communities in
Mexico, Mali, Kenya, Botswana and South Africa to
identify the species that communities find most
useful. Residents of Tsetseng, in the central Kalahari
region of Botswana, are undertaking trial cultivation
of Citrullus lanatus(Thunb.) Matsum. and Nakai
andSchinziophyton rautanenii(Schinz) Radcl.-Sm
in community gardens. In Mexico, MSBP partners
UNAM have identified 339 species used for food in
the Tehuacán–Cuicatlán Valley (Lira et al., 2009) and
are working on the propagation of species such as
Stenocereus stellatus(Pfeiff.) Riccob. In Tharaka,
Kenya, 76 food plants prioritized by local communi-
ties have been collected and conserved at the GeneBank of Kenya and duplicated at the MSB. Associated
ethnobotanical data was collected via a multistage
process, including a pilot survey, questionnaire,
guided group discussions, interviews, transects
walks, observations and photography (Martin, 1995).
This information has been shared with participating
communities via brochures and posters, on-farm
workshops and open days, community tree planting
days, the sponsorship of farmers to share information
during key cultural and medicinal day events in
Kenya’s Eastern Province, and the publication of a
farmer’s guide to seed collection, propagation and
cultivation (Muthoka et al., 2010).5 .Discussion
Seed collections of traditional food plants are of limited
value without the associated knowledge of how to
grow the plants and/or prepare the food product(s).
Likewise, traditional knowledge is of little use if a
community no longer has any seeds or plants of a
particular species. Ex situgene banks should seek
ways to work with ethnobotanists and other social
scientists to complement community based efforts
to conserve traditional food plants. Hawtin (2011)
suggests that the more poorly resourced national
gene banks should focus their efforts on meeting
local needs, rather than attempting to undertake
the whole range of sometimes costly gene bank
activities. Meeting local needs would mean the
maintenance and distribution of materials of imme-
diate interest, including locally important species.
Crucially, materials would be distributed to farmers,
as well as local breeders. Currently, local community
groups may find it difficult to get access to national
(and regional/state) seed collections (Swiderska,
IIED, personal comment).
Hawtin (2011) also suggests that “conserving in-
digenous knowledge” should be a focus for national
gene banks. This will be a challenge. Many gene
banks document only broad categories of plant use- food, medicine, fuel – partly through lack of time
and resources but also perhaps through fear of
112