No abstract

Summary The summarization of large quantities of multivariate data by clusters, undefined a priori, is increasingly practiced, often irrelevantly and unjustifiably. This paper attempts to survey the burgeoning bibliography, restricting itself to published, freely available, references of known provenance. A plethora of definitions of similarity and of cluster are presented. The principles, but not details of implementation, of the many empirical classification techniques currently in use are discussed, and limitations and shortcomings in their development and practice are pointed out. Methods based on well‐defined mathematical formulations of the problem are emphasized, and other ways of summarizing data are suggested as alternatives to classification. The growing tendency to regard numerical taxonomy as a satisfactory alternative to clear thinking is condemned.

No abstract

Textbooks continue to recommend the use of an asymptotic normal distribution to provide an interval estimate for the unknown size, N, of a closed population studied by a mark-recapture experiment or multiple-record system. A likelihood interval approach is proposed and its implementation demonstrated for a range of models for such studies, including all main effect and interaction models for incomplete contingency tables.

Summary1. The decline of the water vole Arvicola terrestris in the UK has been attributed to the spread of the introduced American mink Mustela vison. Understanding the causes and dynamics of this decline is vital to water vole conservation. We investigated the dynamics of water voles in relation to habitat fragmentation and mink predation using an individual-based spatially explicit model of population dynamics on the River Windrush, Oxfordshire, UK. 2. A sensitivity analysis was undertaken using values for life-history parameters drawn from known ranges using Latin hypercube sampling. Partial correlation coecients were used to estimate how the predicted size of water vole population and extinction were determined by the life-history parameters. The model was then validated by comparing model predictions with observed distributions of water voles. 3. The eects of mink predation and habitat fragmentation on the future viability of water vole populations on the River Windrush were analysed after arti®cially manipulating habitat fragmentation on the river and running the model in the presence and absence of mink predation. 4. The match between predicted and observed distributions was signi®cantly related to home range requirement and high reproductive success. At low fragmentation, home range requirement was the most important in¯uence on the number of populations. Reproductive output, and adult and juvenile mortality, became increasingly important with increased fragmentation. At high levels of fragmentation demographic stochasticity had a large in¯uence on population size. 5. We deduce that the importance of demography in determining population persistence will depend on the level of fragmentation. Additionally, life-history parameters that are crucial to the viability of water vole populations can only be identi®ed in the context of the landscape in which populations are found. 6. The extinction of water vole on the River Windrush became more likely as habitat fragmentation and mink predation increased. Mink predation eectively doubled the probability of extinction over that arising from fragmentation alone. 7. These simulations indicate that extant populations on the Windrush are now so fragmented that populations may not be viable even in the absence of mink predation. We assessed the extent of habitat restoration necessary to ensure population persistence on the River Windrush and considered developments of the model for use in water vole conservation.