Dental microwear texture analysis: technical considerations Robert S. Scott a, * , Peter S. Ungar a , Torbjorn S. Bergstrom b , Christopher A. Brown b , Benjamin E. Childs b , Mark F. Teaford c , Alan Walker d a Department of Anthropology, Old Main 330, University of Arkansas, Fayetteville, AR 72701, USA b Surface Metrology Laboratory, Department of Mechanical Engineering, Worcester Polytechnic Institute, Worcester, MA 01609, USA c Center for Functional Anatomy & Evolution, Johns Hopkins University School of Medicine, Baltimore, MD 21205, USA d Departments of Anthropology and Biology, The Pennsylvania State University, University Park, PA 16802, USA Received 30 December 2005; accepted 10 April 2006 Abstract Dental microwear analysis is commonly used to infer aspects of diet in extinct primates. Conventional methods of microwear analysis have usually been limited to two-dimensional imaging studies using a scanning electron microscope and the identification of apparent individual fea- tures. These methods have proved time-consuming and prone to subjectivity and observer error. Here we describe a new methodological approach to microwear: dental microwear texture analysis, based on three-dimensional surface measurements taken using white-light confocal microscopy and scale-sensitive fractal analysis. Surface parameters for complexity, scale of maximum complexity, anisotropy, heterogeneity, and textural fill volume offer repeatable, quantitative characterizations of three-dimensional surfaces, free of observer measurement error. Some results are presented to illustrate how these parameters distinguish extant primates with different diets. In this case, microwear surfaces of Cebus apella and Lophocebus albigena, which consume some harder food items, have higher average values for complexity than do folivores or soft fruit eaters. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Microwear; Diet; Scale-sensitive fractal analysis; Texture; White-light confocal microscope Introduction Dental microwear can provide important evidence for the diets of extinct species. Classically, researchers count and mea- sure individual features on images acquired from tooth surfaces by scanning electron microscopy (SEM) (see Grine et al., 2002 for review). This is a problem for two reasons: 1) identification and measurements of individual wear features are subject to high observer error rates; and 2) the SEM produces two-dimensional representations of 3D surfaces, representations that are depen- dent on the instrument settings and the orientation of the specimen on the stage (see Gordon, 1988). These limitations, and the intrinsic subjectivity they evince, lead to a frustrating lack of repeatability, with high intraobserver (7%) and interobserver (9%) error rates (Grine et al., 2002). Here we describe an observer error-free approach, microwear texture analysis, which yields objective measures of surface texture that do not change with repeated measurement of a given micro- wear surface. Initial work (Ungar et al., 2003; Scott et al., 2005) is summarized, and new efforts are presented. These new efforts offer a more complete characterization of microwear surface textures. Background The many reviews of dental microwear research show the course of progress and resulting advances over the past half century (Teaford, 1988, 1994, 2006; Ungar, 1998; Rose and Ungar, 1998). Early research showed that microscopic patterns of use-wear reflect jaw movements and diet (Butler, 1952; Mills, 1955; Baker et al., 1959; Dahlberg and Kinzey, 1962; Walker, 1976). While these studies were qualitative and relied * Corresponding author. E-mail address: rscott@uark.edu (R.S. Scott). 0047-2484/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jhevol.2006.04.006 Journal of Human Evolution 51 (2006) 339e349