The fundamental fact that teeth wear has been the bane of dental functional anatomists for decades. Researchers have known for a very long time that tooth form reflects function in living mammals, and they have conducted very elegant studies to demonstrate relationships between aspects of occlusal morphology and diet within various mammalian orders. Most such studies have been limited to unworn and slightly worn teeth. The problem with this is that teeth change shape as they wear, and natural selection does not stop when this happens. So how can we infer function from worn teeth? Geographic information systems (GIS) provide one tool to address this issue. In this study, we demonstrate how GIS can be used to assess functional changes in tooth form among variably worn molars of one mammalian species, Gorilla gorilla (Savage and Wyman) 1847.

Most paleontologists who reconstruct fossil mammal diets use the comparative method (Anthony and Kay 1993). They look at relationships between diet and tooth size, shape, or wear in living species to infer diet from the dental remains of fossil forms. There has been a great deal of effort spent on uncovering these relationships. For example, Kay and Hiiemae (1974) associated specific dental morphologies in primates with shearing, crushing, and grinding. Food is sheared between the leading edges of crown crests. Shearing blades generally are reciprocally concave to minimize contact area. In contrast, food is crushed between planar surfaces on the teeth. Grinding involves both shearing and crushing components, where two smooth surfaces are occluded and moved across one another in the manner of a mortar and pestle. Different foods require different actions to break them down for further processing in the stomach.

Kay (1978, 1984) devised a method for measuring the shear potential of a tooth and found that the resultant value, the shearing quotient (SQ), accurately tracks diet in all extant higher-order primate taxa. First, the lengths of mesiodistally running crests connecting the main cusps of lower second molars are summed and regressed over tooth length (in log₁₀ space) for a group of closely related frugivorous species. Frugivorous species alone are used to control for allometric changes in animals that have similar adaptations. The SQs are computed as deviations from the regression line. Thus, positive values indicate longer shearing crests and more occlusal relief than expected of a frugivore and negative values indicate shorter crests and less occlusal relief than expected. For all higher-order primate taxa, folivores and insectivores have higher SQ values than frugivores. Further, among fruit-eaters, those that specialize on hard foods have even lower SQs and blunter teeth than those that more often eat soft fruits (Anthony and Kay 1993; Kay and Covert 1984; Meldrum and Kay 1997). Such studies have served as baselines for paleobiological reconstruction and have provided important clues concerning the diets of fossil primates (e.g., Fleagle et al. 1996; Kay and Simons 1980; Strait 1991; Ungar and Kay 1996; Williams and Covert 1994).

SQ studies have not focused on worn teeth because of difficulties in comparing and measuring crests lengths for variably worn specimens. As a tooth wears, one would expect changing occlusal relief to affect shearing crest length. So how can we best characterize occlusal morphology in worn specimens? This raises one of the most important issues in dental functional anatomy today - how does tooth wear affect function? Because dental morphology affects the magnitude, direction, and rate of change of stress on food particles, tooth shape should reflect the mechanical properties of those foods eaten (Spears and Crompton 1996). How do changes in tooth shape that result from wear affect chewing efficiency? The first step in answering this important question is to develop a technique that can characterize various aspects of occlusal morphology in worn specimens.

Dental topographic analysis is one such technique (Zuccotti et al. 1998). This approach models occlusal surfaces as three-dimensional landscapes using GIS. GIS is an approach used to compare layers of different types of data connected by locations in geographic space. It is a system for assembling, storing, manipulating, analyzing, and displaying geographically referenced information. Many GIS tools have been created to examine and model the physical surface of the Earth. If teeth can be modeled as landscape surfaces, these tools can be applied to provide data on cusp surface area, volume, slope, aspect, and three-dimensional relief. Other functionally relevant aspects of morphology, such as the amount of fluid that could accumulate in a tooth's basin and the directions and intensity of drainage over the occlusal surface, can also be examined with GIS. Such measures may provide insights into food moisture content and other food properties. Walker (1968), for example, noted that pteropodid bats extract juice with low-cusped teeth surrounding a depressed central basin. Studies of occlusal topography are thus likely to be very helpful for reconstructing diets based on biomechanical models of tooth function.

Few researchers have thus far used a GIS approach to the study of tooth shape. Reed (1997) published an abstract describing one technique. He obtained three-dimensional coordinates of small primate teeth with a reflex microscope and interpolated smooth surfaces using Imagine (ERDAS, Inc.). Features were identified with the help of contour lines in ArcInfo (ESRI, Inc.) and relative proportions of area dedicated to cusps, crests, and basins were calculated. Reed suggested that differences in these proportions might reflect diet differences among primates.

Zuccotti et al. (1998) applied GIS techniques to the study of occlusal morphology in great apes. These authors used an electromagnetic digitizer to collect three-dimensional data from each occlusal surface. Resulting data were imported into GRASS (U.S. Army Construction Engineering Laboratory), and tooth surfaces were interpolated using a thin-plate splining model. Cusps were isolated using the lowest elevation contour lines that fully surrounded those cusps, and slope and volume were calculated for each. Drainage patterns and overflow area were also calculated for each tooth (see Materials and Methods section).

Jernvall and Selänne (1999) suggested an alternative approach for smaller teeth (<10 mm in diameter) using a laser confocal microscope. These authors acquired a series of cross-section pictures of the occlusal surfaces of teeth and stacked them together to provide a three-dimensional model of the tooth's surface. Digital elevation models were constructed from the image stacks using the 3D-view version of NIH-Image (U.S. National Institutes of Health), and imported into GIS software (MapFactory, Thinkspace Inc.). Jernvall and Selänne (1999) then demonstrated that various new measurements can be taken. For example, they used the areas of longitudinal and transverse slopes to determine cusp elongation to identify subtle differences in selenodonty in hedgehogs.

As with shearing quotient studies, GIS approaches have thus far focused attention on unworn teeth. The study described here evaluates the potential of dental topographic analysis to document and analyze functionally relevant aspects of occlusal morphology in variably worn teeth of G. gorilla. This study presents a first step toward assessing the functional effects of tooth wear and perhaps even allowing the inclusion of worn teeth in functional analyses that consider occlusal relief in three dimensions. We reconstructed a wear sequence for lower second molars of gorillas using variably worn teeth by scaling and aligning these teeth in an identical manner. We then quantified functionally relevant aspects of morphology for comparison. Results demonstrate the potential of this approach for providing a model of changes in tooth shape with wear. Wear sequence models may be compared among taxa and analyzed for their relevance to tooth function.