What they are. India has a history of religious and cultural traditions that emphasised the protection of nature. In many cultures, tracts of forest were set aside, and all the trees and wildlife within them were venerated and given total protection. These are sacred groves.

Where they are found. In the Khasi and Jaintia Hills in Meghalaya; the Aravalli Hills of Rajasthan; the Western Ghat regions of Karnataka and Maharashtra; and the Sarguja, Chanda and Bastar areas of Madhya Pradesh.

Their role in conservation.

(i) They are a form of in situ conservation. The whole tract is protected, so its biodiversity at all levels is preserved — species, their genetic variation and their interactions — not merely a selected list of organisms.

(ii) They are refuges for rare species. In Meghalaya, the sacred groves are the last refuges for a large number of rare and threatened plants. Some species now survive nowhere else in their region.

(iii) They protect undisturbed reference habitat. Because they have often escaped logging and grazing for centuries, they preserve mature, undisturbed vegetation of a kind that has disappeared from the surrounding landscape.

(iv) Their protection costs the state nothing and is locally enforced. Reverence is a far more durable form of protection than a boundary wall, and it does not depend on the conservation budget — which matters greatly, since the number of species waiting to be saved invariably far exceeds the resources available.

The wider lesson. Conservation is not only a scientific and legal activity. Cultural and religious tradition achieved, in many places for centuries, what legislation has struggled to achieve in decades.

The biotic components act on water before it can do damage — by slowing it, by holding the soil, and by absorbing it.

(i) The canopy intercepts rainfall. Leaves and branches break the force of falling rain, so water reaches the ground gently instead of striking bare soil and dislodging particles. A great part of soil erosion is caused by the impact of raindrops themselves.

(ii) Litter and humus absorb and detain water. The layer of detritus and the humus formed by humification act like a sponge. Being colloidal, humus holds water as well as nutrients, so rain is taken in and released slowly rather than running off at once. This both prevents erosion and converts a sudden flood peak into a steady flow that continues long after the rain stops.

(iii) Roots bind the soil. The root systems of trees, shrubs, herbs and grasses form a dense mesh that physically holds soil particles in place on slopes and river banks. Stratification matters here: a forest with trees, shrubs and a ground layer binds soil at several depths at once, which is why a plantation of one species protects a hillside less well than a natural forest.

(iv) Plants return water to the air. Transpiration removes a large volume of soil water and returns it to the atmosphere, lowering the water table and leaving the soil able to absorb the next rainfall.

(v) Detritivores keep the soil open. Earthworms and other soil animals, in fragmenting detritus, loosen the soil — which is exactly why the earthworm is called the farmer's friend. A loose, open soil takes water in; a compacted one sheds it.

Why this counts as a broadly utilitarian argument. None of these services is bought or sold, yet removing the forest transfers the cost to people downstream in the form of floods, silted reservoirs and lost topsoil. This is the sense in which biodiversity plays a major role in the ecosystem services that nature provides.

In situ conservation is on-site conservation: when we conserve and protect the whole ecosystem, its biodiversity at all levels is protected. We save the entire forest to save the tiger. In India it takes the form of biosphere reserves, national parks and sanctuaries — 14, 90 and 448 of them respectively — together with sacred groves, and globally of the 34 biodiversity hotspots.

Ex situ conservation is off-site conservation: threatened animals and plants are taken out from their natural habitat and placed in a special setting where they can be protected and given special care. Zoological parks, botanical gardens and wildlife safari parks serve this purpose, alongside cryopreservation of gametes, in vitro fertilisation, tissue culture propagation and seed banks.

Why in situ is better where possible.

(i) It protects everything at once. Protecting the ecosystem protects biodiversity at all levels — genetic, species and ecological — including the vast number of species that have never been described and could not be listed for individual protection.

(ii) It preserves the interactions. A species is not only an organism but a set of relationships: its pollinators, its prey, its mycorrhizal partners, its dispersers. Ex situ conservation keeps the organism and loses the relationships, and co-extinction shows how dependent species are on one another.

(iii) It allows evolution to continue. A wild population remains under natural selection and retains its genetic diversity across its distributional range; a captive population loses variation and adapts to captivity.

(iv) It is cheaper per species. One protected forest costs far less than individual programmes for each of its species, which matters because the number of species waiting to be saved far exceeds the resources available.

When ex situ is nevertheless necessary. Where an organism faces a very high risk of extinction in the wild in the near future and needs urgent measures, ex situ conservation is the desirable approach. Many animals now extinct in the wild continue to be maintained in zoological parks — and for them, ex situ conservation is the only conservation there is.

The reason is endemism, not richness alone. The two criteria are very high levels of species richness and a high degree of endemism — species confined to that region and not found anywhere else. Richness explains why there are many species to save in a small area; endemism explains why those species can be saved nowhere else. In a hotspot, losing the habitat means losing the species globally, not locally.

Why the third feature makes it urgent. These hotspots are also regions of accelerated habitat loss. So they combine the highest concentration of irreplaceable species with the fastest rate of destruction — the maximum number of extinctions per hectare lost, and therefore the maximum number prevented per hectare protected.

What the figure 30 per cent actually says. Not that hotspots contain 30 per cent of all species, but that strict protection of them could prevent almost 30 per cent of the extinctions currently in progress. It is a statement about where the losses are happening, not merely about where the species are.

The practical logic behind the whole idea. Faced with the conflict between development and conservation, many nations find it unrealistic and economically not feasible to conserve all their biological wealth, and the number of species waiting to be saved invariably far exceeds the conservation resources available. The hotspot concept is a way of allocating scarce resources for maximum effect. Initially 25 were identified and nine more were added, making 34, of which three — Western Ghats and Sri Lanka, Indo-Burma and Himalaya — cover India's exceptionally high biodiversity regions.

One honest limitation. Concentrating on hotspots means accepting losses elsewhere. Species-poor ecosystems also perform ecosystem services, and a strategy that protects only where richness is highest leaves large parts of the world's biodiversity outside the plan.

The clearest case: pathogens that exist only as human disease agents. The smallpox virus was deliberately eliminated from nature by a global vaccination campaign, and the world is close to doing the same with the poliovirus and the guinea worm. These organisms have no ecological role outside the human body and no other host, so their elimination removes nothing from any ecosystem.

A second case: eradicating an invasive species from a region where it has invaded. The Nile perch in Lake Victoria led eventually to the extinction of an ecologically unique assemblage of more than 200 species of cichlid fish. Removing such an invader — or Lantana, water hyacinth, carrot grass or the illegally introduced African catfish from Indian waters — is deliberate elimination, although note carefully that it is local eradication, not global extinction: the species survives in its native range where it belongs.

How the first case is justified.

On the utilitarian arguments, easily. Smallpox killed hundreds of millions of people and gave nothing in return — no food, fibre, medicine or ecosystem service. Neither the narrowly nor the broadly utilitarian argument for conservation applies to it at all.

On the ethical argument, with more difficulty. The ethical argument holds that every species has an intrinsic value even if it is of no economic value to us. Taken strictly, that would forbid this too. The usual resolution is that our moral duty to the millions of species with whom we share the planet cannot require accepting mass human death from an organism that exists only by causing it — and that the duty to pass on our biological legacy in good order is a duty about ecosystems, not about every genome individually.

Two cautions worth stating. First, this is why smallpox virus stocks are still retained in two secure laboratories rather than destroyed — extinction is irreversible, and the material may be needed for research. Second, the reasoning does not extend to species that merely inconvenience us: mosquitoes transmit malaria but are also pollinators and prey, and eliminating a widespread wild species would be a case of co-extinction risk, not a public health measure.