The Nonconcept of Species Diversity: A Critique and Alternative Parameters

The recent literature on species diversity contains many semantic, conceptual, and technical problems. It is suggested that, as a result of these problems, species diversity has become a meaningless concept, that the term be abandoned, and that ecologists take a more critical approach to species-number relations and rely less on information theoretic and other analogies. As multispecific collections of organisms possess numerous statistical properties which conform to the conventional criteria for diversity indices, such collections are not intrinsically arrangeable in linear order along some diversity scale. Several such properties or "species composition parameters" having straightforward biological interpretations are presented as alternatives to the diversity approach. The two most basic of these are simply ▵₁ =[n/n-1][¹^-^Σ_i (^N _i/_N)² ] =the proportion of potential interindividual encounters which is interspecific (as opposed to intraspecific), assuming every individual in the collection can encounter all other individuals, E(S_n ) = ^Σ_i [1-(^N^-^N_i_n )/(^N_n )] =the expected number of species in a sample of n individuals selected at random from a collection containing N individuals, S species, and N_i individuals in the ith species.