Introduction

Body composition analysis can provide insight into many aspects of an animal's physiology, ecology, and life history. Historically, this analysis has required sacrificing individuals to obtain accurate estimates of body composition. Recent advances in technology and innovative approaches towards developing non-destructive relationships have begun to advance our understanding of body composition without sacrificing individuals. However, these new approaches must still be validated for both accuracy and precision using direct measurements. The purpose of this chapter is to present standard methods for body composition analysis, point out some of the problems encountered when attempting to analyse body composition, and provide guidance for avoiding some of these problems

Statistical issues when measuring body composition

Body composition is a dynamic variable that often shows a high level of variation. When it has been studied, variation in body composition has been found at the level of the individual, population, and species. Although body composition analysis has been used at these three levels, most studies have focused on differences in body composition within a population. Sources of this variation can be both intrinsic (such as age, sex, or reproductive state) and extrinsic (such as climate or resource availability). In addition to natural variation, techniques used to study body composition may also introduce measurement error. Variation, irrespective of the source, must be controlled as much as possible in studies of body composition. Regardless of the taxa being investigated, several steps should be taken to account for natural variation and minimize experimental error.

Sample size

When planning an investigation, the question of an appropriate sample size is crucial.