Principles of Systematics
Determining a "Natural Classification"

Evolutionary processes (anagenesis and cladogenesis) produce a pattern
   phylogeny: the history of organismal evolution
      [cf. genealogy: the history of a single family]
   Diagrams of phylogeny resembles a tree (The Tree of Life)
      living species are the terminal twigs
      extinct species are the interior twigs
      genera, families, orders are successively older & larger (more inclusive) branches & limbs

Systematics: the science of organizing the history of organismal evolution
      the science of ordering

   Identification: recognizing the place of an organisms in an existing classification
      Use of dichotomous keys to identify organisms

   Taxonomy (Nomenclature): assigning scientific names according to legal rules
      Recall discussion of ICZN Green Book (see also Phylocode homepage)

   Classification: determining the evolutionary relationships of organisms
         A "Natural Classification" will accurately reflect phylogeny
            Classification should be a hypothesis of evolutionary relationships

      Alternative classifications are possible (and widely used): But
       An arbitrary classification cannot help us understand evolution
       Ex: If all 'marine mammals' are combined in a single order Cetacea,
                    this implies that aquatic adaptations have evolved only once.
                 If we understand that seals (Pinnipedia),
                    toothed whales (Odontoceti), & baleen whales (Mysticeti) evolved separately,
                    we will understand the differences in their physiology.



Inferring the degree of evolutionary relationship
   How can we describe the position of each 'twig' with respect to all others?
       distance: amount of change between twigs
          How similar (or different) are species?
       phenetic distance: distance measured between tips
                    (i.e., "as the crow flies" from one twig to another)
       patristic distance: distance measured along connecting branches
                    (i.e., "as the ant runs" from one twig to another)
      relationship: pattern of connection between twigs
            How closely related are species?
       cladistic relationship: pattern of branching back to most recent common ancestor (MRCA)
                    (i.e., where do twigs join lower in tree?)

Traditional Taxonomy has emphasized analysis of similarity
Phylogenetic Analysis considers cladistic patterns of common ancestry
    Analysis of distribution of shared character states:
       Character: any morphological, molecular, behavioural, ecological, etc. attribute of an organism
       Character State: alternative forms of a Character [cf. "gene" and "allele"]
    Similarity of characters [character states] may occur for either of two reasons

   Analogous characters are 'similar' because of convergence from dissimilar ancestors
      These do not indicate common ancestry => not useful for classification
         bat wing vs. butterfly wing: embryologically dissimilar
       aquatic habit of whales and pinnipeds
         cow horn vs. deer antler: anatomically dissimilar
         legless lizard vs. snake: common ancestor had legs
       bat wing vs. bird wing: common ancestor was flightless reptile

   Homologous characters are 'similar' because of descent from common ancestor
      These are therefore useful for classification
         bat wing vs. kangaroo arm: from Therapsid forelimb
         ostrich 'wing' vs. penguin 'wing': from Archeopteryx-like wing
       bat forelimb vs. bird forelimb: from reptile forelimb

   Homologous characters will evolve over time =>
   Homologous characters need not look alike or function alike
      Characters that are unchanged from those of the ancestors
         are called 'ancestral' or plesiomorphic
      Characters that are changed in the descendants
         are called 'derived' or apomorphic
      [Avoid the terms 'primitive' and 'advanced': they have false connotations]

      Homologous characters are of two types:
       Shared ancestral characters: similar to each other, and to their ancestor
            also called 'ordinary homologies' or symplesiomorphic characters
                    [This is the usual sense of 'homology" taught in introductory courses]
               Ex.: scales in lizards & crocodiles are an inheritance from Diapsida
       Shared derived characters: similar to each other, and different from their ancestor
            also called 'special homologies' or synapomorphic characters
       Ex.: carnassial pair (P4/M1) is a synapomorphy of dogs & cats
                      is derived from molariform teeth in Creodonta
         [Characters unique to particular taxa are called autapomorphic characters
       Ex.: wings in bats are unique among mammals]

   The nature of homology changes depending on the taxa under analysis
  Ex.: The character "hair" is:
             Among turtle, lizard, bird, and cat: a unique character of mammals
             Among turtle, lizard, cat, and kangaroo: a shared derived character of therian synapsids
             Among kangaroo, bat, cat, and whale: an shared ancestral character of non-cetaceans

    Also: wings are an autapomorphy of the order Chiroptera [they evolved once]
              wings are also a synapomorphy of suborders Mega- & Microchiroptera [they are related]


Homologous characters can be used to construct a natural classification
    (see also short summaries from Ridley 1996; Campbell et al. 1999: Note Homework assignment)

   Use of analogous characters results in polyphyletic groups:
      loosely, groups that do not have a common ancestor
         [but everything has a common ancestor]
      accurately, groups that do not include the common ancestor of the group
       Ex.: Pinnipedia (marine carnivores) were once thought to be polyphyletic
                    walruses & sealions are related to bears,
                    earless ("true") seals are related to weasels
                    [This turns out not to be so]
         Polyphyletic groups are often defined by "absence" characters
        Amphibia: scaleless tetrapods
                The first terrestrial tetrapods (Devonian Amphibia) had scales
                 Modern Lissamphibia [salamanders (Caudata), frogs (Anura), & caecilians (Gymnophiona)]
                        are secondarily scaleless [an adaptation for dermal respiration]
                        & probably  independent lineages
        Edentata: toothless mammals
                    Jurassic mammals had teeth
                   anteaters (Xenarthra) and pangolins (Pholidota) are secondarily toothless

   Polyphyletic groups are rejected by all modern taxonomists
         They do not have 'evolutionary implications'
                 'edentate' [toothless] taxa evolved under distinct ecological conditions
       Ex.: "Insectivora" is a  'garbage can'  taxon:
                any "primitive" insect-eating animal that doesn't fit elsewhere

   Use of homologous characters results in monophyletic groups:
      loosely, groups that are descended from a single common ancestor
      accurately, groups that include the common ancestor of the group

   Monophyletic groups are of two kinds:

   Use of shared ancestral characters results in paraphyletic groups:
      a monophyletic group that includes the ancestor and
      some but not all of its descendants. This creates a

      Grade: a group defined by a combination of shared ancestral & derived characters
         describes a level of biological organization

   Ex.: among the traditional taxonomic Classes of Vertebrata
     Agnatha: jawless descendants of first vertebrates
            hagfish (Myxiniformes) & lampreys (Petromyzontiformes)
         gnathostomous relatives of Craniata (Chondrichthyes, "fish") not included
       Osteicthyes: fish with bony skeletons
         amniotic relatives of Sarcopterygia (lungfish) not included
       Reptilia: scaly tetrapod descendants of first amniote
         feathery diapsid & furry synapsid relatives not included

Paraphyletic groups are accepted by traditional ("Evolutionary") taxonomists,
                                       and rejected by 'Phylogenetic' taxonomists ("Cladists").

Phylogenetic taxonomists make the following arguments

  Classification should reflect only relationship, not similarity
  Relationship can be determined objectively, e.g., by molecular methods
            Organismal similarity & differences are what we are trying to explain

   Grades are subjective
  Which character is more important?
            'Scaly' reptiles are what's left when you take out  'feathery' birds and  'hairy' mammals
            Why not take out "finny" Icthyosauria (marine reptiles)?

   Grades perpetuate biological & evolutionary myths
            "Reptiles & lungfish aren't very variable.
                Their body plans have limited evolutionary possibilities."
            "Dinosaurs are more like reptiles than they are like birds."

   *** Grades are not useful units for evolutionary analysis ***

    Ex.: Evolution of pagophilic (ice-breeding) behavior in phocid seals (Perry et al. 1995; Carr & Perry 1997)
       Phoca vitulina (harbour seals) breed on land,
                 other seals (e.g., Phoca groenlandica & Halichoerus grypus) breed & nurse young on ice
       Traditional taxonomy suggests ice-breeding has evolved several times:
                     Separate explanations for each pagophilic species are required:
                 e.g., ice-breeding is a polar bear avoidance behaviour
       Phylogenetic taxonomy (supported by molecular analysis) suggests ice-breeding is ancestral:
         Phoca groenlandica is a separate genus Pagophilus groenlandicus,
                     & is more closely related to ice-breeding seals like Cystophora
                  => Phoca shows a recent evolutionary shift to terrestrial breeding,
                     special explanations for ancestral pagophilic behaviour are not required.


  Exclusive use of shared derived characters results in holophyletic groups:
      a monophyletic group that includes the ancestor and all of its descendants
      Clade: a group defined by one or more shared derived characters
         describes a complete ancestor-descendant lineage

   Ex.: Among the traditional classes of Vertebrata
           * Placodermi: gnathostomes with a hinged craniovertebral joint in skull (extinct)
           Chondricthyes: gnathostomes with a hyostylic jaw suspension
           Mammalia: cynodont therapsids with a dentary-squamosal jaw suspension & hair
           Aves: archosaurs with feathers
           Saurischia: amniotes with a diapsid skull includes both Mammalia & Aves

      Holophyletic groups are accepted by all modern taxonomists,
       'Phylogenetic' taxonomists use them exclusively [but some don't like the term]
         'Traditional' taxonomists regard reliance solely on clades as misguided

Evolutionary taxonomists make the following arguments:

   Classification should reflect similarity as well as ancestry
        It is commonly understood that turtles, lizards, & crocodiles are more similar than different,
                    and are quite distinct from birds
       [Is this true? Depends on the features examined]

           Clades obscure biological distinctiveness
       Inclusion of Birds as Dinosauria obscures their differences
               [Are they that different? Note hip structure & bipedalism.]
                Since "Jurassic Park": dinosaurs are "more like" birds: NOT

   Clades overemphasize minor differences
               Classification of amniotes by temporal openings "over-splits" group
                  [Are these differences so minor?]

   Classification by clade leads to unfamiliar names
       Gnathostomata, Amniota, Sauropsida, etc.
                    [But these correspond to major evolutionary innovations:
                        Jaws, amniotic egg, & water-impermeable skin
       "Need to Name" every branch leads to proliferation of categories
                (legion, tribe, cohort, etc.)
                    & prefixes (super-, sub-, infra-)
                    & suffixes (-oidea, -idae, -inae, -ini)
                        [for superfamilies, families, subfamilies, tribes, respectively]
                     [Is this necessarily bad?]

       Ex.: In the traditional taxonomy of Primates
                      Hominidae (Homo) is separated from Pongidae (Pan, Gorilla, Pongo) (great apes)
                       => perceived similarity of apes & distinctiveness of Homo are emphasized,
                            relationship of Pan & Homo is obscured
               In a cladistic taxonomy of Primates
                       Homo, Pan, & Gorilla grouped as Homininae
                      Homo & Pan grouped as Homini (or Panini)
                       relationship is emphasized
               Does this obscure ape similarities?


Text material © 2010 by Steven M. Carr