Tuesday, July 13, 2010


A way to incorporate time into the calculations, is through the technique of "stratocladistics" (rather than "cladistics").

"Stratocladistic hypotheses attempt to explain both the distribution of characters among taxa and the distribution of taxa through time. By doing this, stratocladistic hypotheses explain more features of the natural world and hence have greater explanatory power than purely cladistic hypotheses."

"Stratocladistics is a method of making phylogenetic inferences using both geological and morphobiological data. It follows many of the same rules as cladistics, using Bayesian logic to quantify how good a phylogenetic hypothesis is in terms of debt and parsimony. However, in addition to the morphological debt that is used to determine phylogenetic dissimilarities in cladistics, there is also stratigraphic debt which adds the dimension of time to the equation."

"Debate has long simmered over whether data on the order of appearance of taxa in the stratigraphic record should play any role in analyses of phylogenetic relationships among those taxa. Critics argue that temporal data are in principle inapplicable to questions of cladistic relationship, but specific versions of this claim all seem flawed. Stratocladistics offers a methodological context (patterned after that of cladistics itself) within which temporal data participate along with conventional character data in selecting most-parsimonious hypotheses. Stratocladistics outperforms cladistics in tests based on simulated histories, and additional testing will be facilitated by new software automating stratocladistic searches. As with any body of data, we may decide to include or exclude temporal data for specific reasons, but the explanatory power of hypotheses that use both temporal and conventional character data exceeds that of hypotheses based on character data alone." (Daniel Fisher)

No comments:

Post a Comment