GEOS 26400/36400/BIOL 23255/EVOl 32300: Principles of Paleontology

Systematics, II & III
I. Background to Phylogenetics and Classification
 -Uses of phylogenetic data (just a small sampling)
	-evolutionary rates
	-evolutionary trends, preferred directions of change
	-coevolution
	-heterochrony and other evolutionary patterns, mechanisms
	-"tempo and mode"
	-macroevolution, hierarchical approaches
	-biological diversity
	-correlates (causes) of diversification
	-nature and temporal & spatial deployment of evolutionary novelties
	-origination and extinction

  -Nature and purpose of classification
      -classification as system for information retrieval
      -classification as summary of morphological similarities and differences
      -classification as summary of ecological similarities and differences
      -classification as summary of evolutionary relationships
      -Purposes may be at odds!
      -Formal taxonomic decision always involves two distinct elements:
	1. inclusiveness (which organisms or lower taxa are contained)
	2. rank (level in the taxonomic hierarchy)
      	  -Kingdom, phylum, class, order, family, genus, species etc.
          -subjective element to ranking
          -Two taxa of higher rank generally are morphologically more
                  dissimilar from each other than two taxa of lower rank.
          -Criteria that go into taxonomic rank include diversity and 
		morphological distinctiveness
   -Origin of higher taxa
           -Higher taxa may be distinct when they first appear...
           -...or they may be recognized as higher taxa only in hindsight,
             because they have accumulated diversity and divergence over time.
   -Cladograms versus trees
	-cladograms: topology of branching
	-trees: topology as well as time and evolutionary events

II. Monophyly, paraphyly, polyphyly
      -general agreement that polyphyletic groups are undesirable
      -disagreement about paraphyletic groups
            *Debate concerns classification, not evolution!
            -Exclusion of descendants seems arbitrary to some.
            -But such exclusion may allow emphasis of differences deemed important (e.g. 
		ecologically, physiologically)
		-example: "amphibians" versus amniotes
            -Paraphyletic groups sometimes exaggerate extinction events, sometimes not
            -Insistence on holophyletic groups makes classification of fossil taxa difficult.
		-stem-crown-plesion concept problematic with very diverse fossil groups

III. Some basics of phylogenetic analysis
      -homologous vs. analogous characters
      -primitive vs. derived characters
      -shared primitive vs. shared derived characters
            -shared derived characters as key to branching sequence
      -cladistic grouping (ideally) based on shared derived characters
      **Discussions of phylogenetics often emphasize data analysis, but character analysis 
that leads to data matrix is the crucial step.

IV. Overview of cladistic parsimony (cladistics)
	-Caution: Underlying assumption is not that evolution acts parsimoniously.
	-Rather, parsimony used as scientific operating procedure 
(in part) because we generally infer evolutionary transitions indirectly 
rather than observing them directly, and therefore we want to minimize the 
number of such ad hoc inferences we must make.
	-Considerations of parsimony should be downweighted if there is 
independent evidence (i.e. independent of the character data at hand) of 
convergence or reversal.
		-e.g. laminate leaves
		-e.g. stemlessness in crinoids
	-Approaches to establishing character polarity
		-outgroup comparison
		-stratigraphic position
		-developmental sequence
	-Unrooted analysis of unpolarized characters
		-followed by designation of root to polarize characters

V. Long-branch attraction
   
	*If evolutionary rates vary among branches, then branches tend to group together because they have similar rates rather than because they are closely related (higher probability of independent convergence of novel character).
	*Common solution is to add taxa to analysis to "break up" the branches.

VI. Maximum likelihood and related methods

	*General approach: Determine probability of observing given character data under assumed model of evolution and postulated evolutionary tree.  Tree with highest probability is the 
maximum-likelihood estimate (for the given evolutionary model).
	*Problem: Reasonable models may be difficult to formulate, 
especially for morphological (as opposed to molecular) data.
   
VII. Role of temporal (stratigraphic) information
      -Many workers ignore age of fossils.
      -But there's a known correspondence between inferred branching sequence
            (clade rank) and order of appearance in the fossil record.
      -Estimation of evolutionary trees requires incorporation of temporal data.
      -At least four ways to incorporate time into phylogenetic analysis
        1. -Stratophenetics: link species based on overall similarity and stratigraphic 
		position.  (Return to this later in discussion of punctuated equilibrium.)
	2. Use stratigraphic data to sort among equally parsimonious cladograms based 
		on their consistency with stratigraphy.
	3. Use stratigraphic data plus character data to recognize ancestors
        4. Stratocladistics: maximize the joint parsimony of morphological characters and 
		stratigraphic data. 

VIII. Role of fossil taxa
      -Some studies show that inferred relationships among living taxa are
            affected by inclusion of related fossil taxa; others disagree.
            **Fossil taxa especially useful when living representatives are highly derived.
	    **Fossil taxa essential in inferring ancestral character states and actual
		patterns of evolutionary modification.  (E.g. fact that a living fly and a spider 		are more closely related to each other than either is to a cow tells us very 			little about the common ancestor of the two arthropods and the evolutionary 			changes that led up to these living forms.)