The fossil record of the Paleozoic Bivalvia can be described in terms of directional and repetitive evolutionary trends. The directional aspect is reflected in long-term generic diversification, and in increased abundance and dominance of cosmopolitan genera. These changes are believed to have resulted from an increase in the degree of environmental variability in shallow-marine, continental margin regimes.
Repetitive evolution is recorded in the temporal-geographic pattern of speciation and extinction; it appears that when shallow-marine environments were disrupted periodically, presumably by some major physical event, the bivalve populations responded in much the same way each time.
Thus, given measurements of bivalve geographic and stratigraphic distribution, evolutionary hypotheses can be formulated:
Hypothesis 1. Fluctuations in total standing generic diversity in an ecologically and phylogenetically coherent group of organisms undergoing long-term diversification describe a pattern in which numerical differences between the successive levels of high diversity are related by a constant multiplier.
Hypothesis 2. Following a major extinction event, faunal replacement occurred in two distinct waves separated by a period of little faunal change—the second replacement wave immediately preceding the next extinction event.
Hypothesis 3. Patterns of speciation and extinction are reflected in repetitious endemic-cosmopolitan interactions. Major extinction events, probably the result of a continual deterioration of the physical environment, progressively affect more widely distributed populations as the physical event approaches a climax. Thus populations with a history of cosmopolitanism have a high probability of surviving as stock populations that give rise to subsequent diversifications, the first of which also tends to be predominantly cosmopolitan. After this initial radiation, there is a period of low faunal turnover, followed by another radiation dominated by endemics, which, in turn, are susceptible to the factors responsible for the next extinction event.
Hypothesis 4. Those higher taxa (for example, superfamilies, orders) whose earliest representatives (genera) tend to be highly endemic and which retain this endemism for a few epochs are less long-lived than higher taxa that initially have a cosmopolitan representation, albeit a numerically sparse one.
Many paleontological studies have documented directional patterns in evolution—changes in size or in morphological complexity within a given taxon, or increase in biotic diversity throughout Phanerozoic time—but the recognition of repetitive patterns in the fossil record may stimulate the eventual development of predictive theories, especially regarding the times of the conditions for mass extinctions. Both types of evolutionary patterns require the development of hypotheses that can be tested and conditionalized utilizing data other than those presented herein.