The world must act far more urgently to save thousands of threatened wild plants, three British botanists say.
They say part of the answer is seed banks, which can be richer in higher plants than any location on Earth.
The cost would be a tiny fraction of the amount spent on particle physics, and would pay huge dividends, they say.
Even seeds from rainforest plants and trees can be saved in banks, some of them for centuries, provided the conditions are carefully controlled.
The argument for seed banks is made in Biologist, the journal of the Institute of Biology, by three scientists from the UK's Royal Botanic Gardens, housed at Kew Gardens in west London.
The three, Simon Linington, Clare Tenner and Roger Smith, are all involved in the development of the Millennium Seed Bank Project at Wakehurst Place, Kew's country garden in southern England.
They say many seeds probably have the potential to stay viable for two centuries, so seed banks can be significant in conserving genetic wealth for future generations.
The need for conservation is stark. At least 34,000 plant species are globally threatened, the authors say, and they cite a 2002 estimate by the eminent US scientist E O Wilson, who said as many as 50% of plant and animal species could be on the brink of extinction by 2100.
Threats include land use changes, habitat loss, invasive alien species, over-exploitation, and climate change.
But our knowledge of how seeds will cope with conservation is relatively slight. The behaviour of seeds in storage is known for only 4% of angiosperms (flowering plants).
Of the species examined, 92.4% have seeds which will tolerate drying and can therefore be banked.
The seeds of 5.9% of species (including, for example, most oaks) die if they are dried below about 40% moisture content, while the rest can be dried to about 10%.
The authors write: "Despite assertions to the contrary, many tropical and even rainforest species can be banked."
The contribution seed banks can make is therefore huge, and some contain more higher plant species per square metre than anywhere else on the planet.
The development of seed banks has led to disputes over who owns their botanical resources and who should have access to them.
The United Nations Convention on Biological Diversity has clarified some questions of ownership and benefit-sharing.
But the authors are disappointed that some key crops remain outside the provisions of the International Treaty for Plant Genetic Resources for Food and Agriculture.
They write: "Consequently, African plant breeders will have no easier access to groundnut diversity from Latin America as a result of this treaty (there is tremendous interdependence on crop resources between countries)."
The adoption by the convention's members of the global strategy for plant conservation, they say, was "an important milestone".
Implementing the strategy's target that 60% of threatened plant species should be held in accessible collections, preferably in the country of origin, and 10% of them included in recovery and restoration programmes, would cost about $100m, they say.
A further $5m annually would be needed to maintain the seed banks.
They add: "The initial outlay is just 2% of the $6bn that Cern (the European Laboratory for Particle Physics) and the US agreed in 1997 to spend on the Large Hadron Collider" (a massive atom-smasher).
"To achieve their goals, plant conservationists should learn from the physicists' political lobbying skills."