Studying Sex In Fossils: Some Caveats
Problems due to Quality of Measurements -- Quality of Materials Problems due to Taxonomic Status -- Geological Site -- Investigator The Problem of not Knowing which Sex is which The Effects of Taphonomy General Problems of Multivariate Statistical Analyses Biological Interpretations -- Statistical Validity Research Design -- Replicability The Special Problem of Fossils in Multivariate Analyses
Abstract: The work outlined in previous chapters depends upon the use of particular types of data and particular statistical methods. The results seem clear; the methods, though somewhat complicated, have been worked out by statisticians and applied in many biological situations. Yet, in every investigation both data and methods are subject to constraints that need to be understood before discussions and conclusions can be broached. We must be appropriately self-critical and try to understand the qualifications to which this work, indeed all studies of this type, are subject. Of all things, this, self-criticism, is the most difficult to carry out.
All of these studies have been carried out using the overall lengths and breadths of teeth. Surely, says the percipient reader, something better than overall dimensions could have been measured. And this comment is indeed appropriate. There are many studies in the literature, wherein the complicated forms of teeth have been better captured through the use of other data.
Sometimes this has encompassed, through observation or measurement, the pattern of placement of the various complex cusps of the teeth. Study of these patterns has resulted in much of what we know about many fossil groups. They are often enormously complicated and contain a great deal of information about the relationships of animals. It has been shown many times that such patterns are somewhat akin to finger prints and reflect the underlying genetic make-up of animals. For instance, the incidence of 'Carabelli's cusp' is important in understanding human genealogies. Cusp patterns also tell us something about animal diets. We are all aware, for example, of the different cuspal patterns in carnivores and herbivores, and of the especial complexities of cusps in animals, such as elephants, reflecting masticatory processing of large amounts of very tough foods.
Sometimes, too, measures have been specifically aimed at defining the cutting, shearing and crushing edges and areas of the teeth. This is clearly related in part to the cusp patterns (though not completely, for cusp patterns may be eliminated by tooth wear during use, but shearing, cutting and grinding edges and surfaces remain; perhaps original cuspal patterns help determine eventual wear patterns). Such patterns, even more than those of the cusps, are related in a specific manner to different masticatory habits, and different food materials. We can look to carnivores and herbivores for extreme specializations along such lines.
Dental areas are also associated with other facets of the lives of animals. They are related to their overall physiological and metabolic requirements. They provide measures that can be seen in the light of overall amounts of food materials being triturated, overall energy requirements being met by the food, overall nutritional values of particular types of food materials, and so on.
And yet other measures of tooth areas, for example, cross-sectional areas of the main body of the tooth are related to the totality of stresses borne