EVOLUTIONARY DECELERATION: THE ROLES OF CLADOGENESIS AND ANAGENESIS

Morphological studies of many different clades have documented a pattern of evolutionary deceleration where early clade history is characterized by more rapid morphological diversification than later clade history. Although the pattern of evolutionary deceleration seems to be wide spread, the relative roles of anagenesis and cladogenesis in causing this trend are poorly understood. This study calculates rates of evolution using character state information and biostratigraphic data for 53 published phylogenies of different groups. To overcome the limitation imposed by the small sample of rates from any individual group, two different approaches that utilized data from all 53 phylogenies were used to assess the trends in rates: the binomial distribution test and a normalized aggregate analysis. For the binomial test, the number of individual groups with decreasing rates through clade history was significantly more than the number with increasing rates. The aggregate analysis, which used a normalized scale for both rates and time so data from all groups could be combined, also showed a significant pattern of morphological deceleration through clade time. This trend is significant both for invertebrates and vertebrates and it withstands several sensitivity analyses.

The number of character state transitions per node as well as the rate of speciation (nodes/my) decrease suggesting both anagenesis and cladogenesis are driving the observed pattern of evolutionary deceleration (evolutionary steps/my). The influence of cladogenesis appears to vary throughout geologic time in relation to external factors like open ecospace. For instance, higher rates of morphological evolution during the recovery phases after major mass extinctions (e.g. K-T, P-T boundaries) are driven mostly by higher rates of cladogenesis. Higher rates of anagenesis, however, are not as closely associated with mass extinctions and thus seem to be mostly constrained by internal clade dynamics.