| [Thread Prev] | [Thread Next] | [Thread Index] | [Date Prev] | [Date Next] | [Date Index] |
Peter Sheehan writes: >My guess is that your son will be more interested in how families of >horses or graptolites lived rather than how many species were in the >family. If the study of life on earth was just a matter of figuring out >how many species were present now or in the past I would get out of the >business. There it goes again -- my own rhetorical devices (in this case, hypothetical sonny boy) getting me into trouble. No question that how ancient organisms got along and what adaptations they used to do it are rich lines of inquiry. Queries about timing and extent of diversification are valid too. What my little trope may have failed to get across is my doubt that a bean-counting style of systematics can take us much farther than it has already. No paleontologist I know would ask the question "how many species make a family?" with a straight face. At some point in our education we learned not to. I can't recall the text; in my case it was probably just one of those things I soaked up. A family is a grouping of genera -- what other definition is there? What do the taxa Helminthoglyptidae, Hominidae, and Rosaceae have in common that makes them all "families?" Then Neil Clark writes: >I would add that there is no need to have a constant interpretation as >to what a Family represents in terms of its relationship with other >Families. Especially as we are not sure even whether species-level >taxonomy is equivalent in either living or fossil organisms. This one confuses me. In a classification, two families would represent two mutually exclusive groups of genera. But in a phylogenetic analysis, one "family" of canonical systematics may turn out to be a subset of another "family." Is it all right for both "families" to carry equal weight in an analysis like Benton's? Barry Roth barryr@ucmp1.berkeley.edu Research Associate, Museum of Paleontology University of California, Berkeley, CA 94117 USA (415) 387-8538
Partial index: