Predicting and understanding rarity: the comparative approach

Abstract

Over the next several decades, actions will need to be taken towards preserving the many species now facing extinction. Unfortunately, this will be done in the context of enormous ignorance about most of these species. G. Mace, 1995 13.1 INTRODUCTION Our ignorance about the world's organisms is astonishing. For many species, we do not even know whether they are rare or common. For others, the fact that they are rare is the most significant piece of information we have. In a small number of cases, we know how rare a species is and also have a good knowledge of its life history, ecology or behaviour. When we have this information about a whole group of species, we can ask what categories of species tend to be rare, merely by tallying the characteristics of known rare species. Such tallies have their uses but, for two reasons, they may not necessarily reveal what features of the species are the cause of their rarity. First, variables that are correlated with rarity could equally well be causes or consequences of that rarity and we must be careful in deciding between these alternatives. Second, correlations found between two or more variables that are measured across species may be coincidental artefacts of the methods used and not true evolutionary correlations, by which we mean that the two characters have evolved together on a number of independent occasions. The first problem is undoubtedly complex and difficult but it can occasionally be overcome. Towards the end of the chapter we will demonstrate how. However, this chapter is mainly about solving the second problem -how to identify evolutionary correlations. The techniques of studying evolutionary correlations across species are those of the comparative method (Ridley, 1983; Pagel and Harvey, 1988; Harvey and Pagel, 1991) and we shall explore a variety of these methods in this chapter. We begin with some simple, exploratory statistics, useful for making predictions about rarity, before considering newer, more sophisticated methods for recognizing ways in which rarity has co-varied with other features over evolutionary time. We use different measures of rarity -range size, population density, endangeredness -in different examples, merely to demonstrate the versatility of the methods. 13.2 THE BIRDS OF AUSTRALIA