There are very different numbers of species in different parts of the Earth: the highest species richness on land occurs in the tropics, and specifically in the tropical rain forests. These forests occupy only about 6% of the Earth's surface, and may have more than half of all species. A possible cause of this difference seemed to be the fact that more energy from the sun reaches the tropics, thus more food-plants can grow, and everything else lives on these plants. Tropical coral reefs are the most species-rich shallow marine regions. Many of these corals do hardly eat, and live dominantly on the energy that they receive from small algae that live inside their bodies - and the same argument of abundant sunlight might be valid for these reefs that live on their photosynthesizing algae.

But just having more energy does not automatically imply more species: more energy means more biomass, but not necessarily biomass in more species. Some of the most productive areas in the world are coastal salt marshes, which have huge amounts of biomass and very few species of plants. Energy can not be the whole story: in the deep oceans it is always cold and dark, and just a very little bit of food arrives, falling down from the sun-lit surface thousands of meters above. The deep oceans have an incredibly great species richness, and at hardly any energy available.

So what do these three very different regions with high species richness have in common? A good description of the reasons for this difference in species richness by latitude comes from Colinvaux's book 'Why big fierce animals are rare': life at higher latitudes on land, at shallower depths in the oceans, tends to be 'accident-prone'. In these regions with low numbers of species the physical environment shows great variability, for instance, as a result of the changing seasons. No organism can be a specialist, all organisms must be able to live through the varying seasons (or migrate), and at higher latitudes the seasonal differences become more extreme (e.g., total dark during the winter, freezing). The non-specialist organisms from higher latitudes survive over rather large areas - large areas means varying conditions, and they can deal with that. The organisms in such environments, however, are prone to becoming extinct: at high latitudes the summer is short, and one summer of bad weather can kill off huge populations in little time. We can thus argue that chances of going extinct are higher at high latitudes than at low latitudes: the gradient in species richness (from high in the tropics to low at the poles) reflects a gradient in chances of becoming extinct. The few generalist species that survive can form large populations and thus become less easily extinct (see below).

It has long been thought that more species-rich ecosystems are more stable, that is, can less easily be disturbed, than less species rich ecosystems. This appears logical: assume a very simple ecosystem, with plants, rabbits and foxes. If many rabbits die as the result of a freak accident, the foxes will be in difficulties because too many of them will try to eat too few rabbits and many will starve. If the foxes die out by a freak accident, the rabbits are badly off because they will overgraze the plants and many of them will starve. This system is very unstable. But if we imagine a system with 10 species of rodents and 10 species of carnivores, the unlucky extinction of one species might not do much harm - there's other choices for the survivors. This was called the complexity-stability theory, which is no longer widely accepted. Note that we just argued (above) that climatic stability results in a high species diversity. We are doing some circular reasoning if we now say that high diversity causes stability in ecosystems. Might it not be so that the high diversity in tropical areas is caused by the climatic stability?

One the reasons of the early popularity of the complexity-stability hypothesis was the fact that ecologists could use mathematics developed for information theory, in this case, the way in which information flows through various channels (developed for phone lines, by Bell Labs). Obviously, more possible phone lines will mean a more stable system, because if one line breaks other can take the messages. It was then argued that in nature energy flows through food webs similar to the flow of information through phone lines, and that similar mathematical systems could be used. The problem is that in phone lines the message does not care what lines it goes through, but in ecosystems the various species are not fully interchangeable: animals eat different things, for instance, and are not able to suddenly switch diet.

It is now though that the highly diverse tropical ecosystems are much more vulnerable to physical disturbance than the lower diversity high-latitude ecosystems: their organisms are not used to variable conditions, and can not survive environmental changes easily. The numerous species include many specialists that live in small regions, and consist of few individuals, and small populations can easily be wiped out by a local disaster. In addition, the small populations may decrease to such a small group that inbreeding occurs and diseases become common and lead to extinction. Higher latitude, less diverse ecosystems, however, are used to much environmental change and are able to survive much more severe disturbances.