Art Weil asked how estimates of extinction rates are made. This is a good question. Some groups are relatively well known, such as birds. Most birds are active during the day, and there are hordes of amateur and professional ornithologists to study them. It is very unlikely that any new species of birds will be discovered in North America this year; we have an accurate species count. We also know which birds have historically become extinct, and in some cases we even know the day when they became extinct, so we have a good idea of how many species of birds became extinct per century, or per decade, in North America. The situation is similar in other parts of the world, though new birds are still being discovered in places like New Guinea. These numbers can be graphed and projected into the future. That is one source of statistics on extinction rates. Other groups are not quite as well known as the birds. Mammals are mostly active at night, and new species of bats and mice are still discovered occasionally. But it is only in very remote places that new species of large mammals can still be discovered, e.g., the jungles of southeastern Asia. The same statistical methods can be applied to mammals as for birds, but with a greater possibility of error. If the discovery rate for mammal species is plotted on a graph, it shows a steep rise early on, then it flattens out as almost all mammals have been discovered. A similar situation might be demonstrated for freshwater mussels. We have a rough idea of how many species are threatened or endangered in North America, and how many have been lost in the last few decades, and where. If a knowledgeable person says, "We've lost (say) one-third of the species of endemic land snails in Hawaii and are likely to lose another third in the next few years," it is an estimate based on hard statistics. (Incidentally, I just made up the 1/3 figure for purposes of discussion; I don't know the actual numbers for Hawaii.) Some groups are not well known at all. New species of insects are named by the thousands every year, particularly from the tropics. If the discovery rate for insect species is graphed, it doesn't show any sign of slowing down at all. Evidently, there are a LOT more insects out there to be described for the first time. And marine mollusks too, I might add. How can we estimate their number? This isn't easy. One method is to list all the species known from an area, then sample the area intensively and see what proportion is new. Then correct the total number of insect species in a larger region by the same proportion. The estimated numbers rose steeply a few years ago when an enterprising biologist wrapped whole trees in plastic and gassed them to kill every insect. The number of new species living in the tree canopy turned out to be phenomenally high. Biologists who formerly estimated the total number of insect species as 1 or 2 million now talk about 10 or 20 million. The range of error is correspondingly great. Now, it is known that much of the world's diversity lies in the tropical forests. It is also known that many plants and animals live only in small parts of this forest, or live so widely scattered that they require vast stretches of forest (compare the peregrine falcon of North America, for instance). It is usual for a square kilometer of land to have no two trees alike, if I remember right, a situation that is absolutely amazing to Europeans and North Americans. And other organisms, from insects to fungi, often live on only one part of one species of tree, so this system is highly vulnerable to species loss if large tracts are cut down. Well, large tracts are being cut and burned, and much of this land is being turned into farmland and pasture instead of returning to forest. The idea among biologists is that if half the forest is destroyed, then more than half of its species will be destroyed. And that's why they say things like, "Thousands of species will be destroyed forever before we even have a chance to see them for the first time," which admittedly sounds strange at first! And that's where you get some of the latest estimates of extinction rates. And that is why Kurt Auffenberg is haunted by the idea that so many species of land snails will vanish without being recorded, and why he makes it his business to collect and record as much as he can. That is why Marlo Krisberg has been trying to document the species of one area of Florida while collecting very little. That is why Doug Shelton spends some of his own time to look for species of freshwater mussels that may not be extinct yet and will need protection to survive, photographing mussels and returning them to their bed. What can be done? The quality of the environment can be improved: less pollution, more effective ways of production, more efficient use of materials per person. Refuges can be set aside for whole forests and reefs. Zoos and botanical gardens can save a few plants and animals. Seed banks can be maintained, especially for the most important agricultural species. Expeditions can be financed to collect and record the organisms that live in threatened areas, and comprehensive monographs can be compiled on the organisms of particular regions. The dispersal of organisms to new habitats can be discouraged. Individuals can contribute money, or grow rare organisms in greenhouses, or document the fauna of their favorite area, or vote. But in the long run, there will have to be fewer people, or we will have a lot fewer kinds of plants and animals on this planet, and that does include shells. We are still in the beginning of this great crisis of extinction, and we can make a difference if we act now. Later will be too late. Andrew K. Rindsberg Geological Survey of Alabama