Of late there are several newly compiled databases giving the stratigraphic ranges of extinct and surviving groups being made available on the Internet. The most comprehensive of these is The Fossil Record 2 (Benton 1993) for which experts in each phylum of animals and plants estimate the time of first and last appearance of a total of 7,186 families. Although this database is a great advance over the first version (published 25 years earlier), there remains substantial distrust of some of its content. This uncertainty has impeded the development of reliable mathematical models describing the diversification of these organismal groups. As long as these data depend on the fossil record no diversification model can be believed to be absolutely correct; there may always be the possibility that some maxima or minima are artifacts of uneven fossilisation processes, the effects of an incomplete record, or human errors in collecting the data, identifications, and database construction. Nevertheless, the well-established theory of punctuated equilibrium (Eldredge and Gould 1972) and the familiar family diversity curve (Sepkoski 1984) are relied upon by many evolutionary biologists. Their favoured step-wise model of diversification has been supported recently by Courtillot and Gardemar (1996), but our own analyses give a very different interpretation (Hewzulla et al 1999).
Despite the effect of these well-known aspects of the fossil record, diversification patterns may gain more confidence by comparison with different sources of evidence, such as geology and molecular biology. For example, if some maxima or minima coincide with the times of continent collisions such patterns may more closely reflect the actual process of biological diversification. The first part of this study was undertaken in order to examine the effect of interdatabase comparisons in estimating the accuracy of patterns generated from databased fossil stratigraphical ranges. Because there are so many of these now available on the world wide web (WWW), we present a technique that enables any user to interact with a web-based database and have complete choice in selecting different kinds of data to produce different diversity curves. If you have a Java-enabled browser, you can make visual comparisons of whatever you choose from The Fossil Record 2 database.
Our investigation then goes on to present a mathematical model based on the assumption that the diversification pattern of any organismal group fluctuates around a global trajectory defined by the Nee and May (1997) mathematical model so long as it has not been exposed to strong external environmental perturbations. The Nee and May model is a modified logistic model that includes a factor controlling the global extinction of the constituent group; in other words, based on the assumption that every organismal group--whether it is a species, family or order--will eventually become extinct, regardless of the diversification rate. Although we accept that the parameters of such a model may vary according to uncertainties in the fossil record of a group, our trials show that the predicted bell-shape the diversification pattern does stay relatively stable.