Phylogeny and classification

 

Lecture notes of Claus Hedegaard, May 11-12 1999

 

The complete hand-outs (glossary and list of literature) are included as overheads 11-12.

 

Topic/text

Overhead

Hand out demonstration problem from text designed for American teachers: 'Biological Classification - organizing organisms', http://www.biology.aau.dk/genetic.eco/bioch/UniMaine/ Made for Geological Society (USA)

 

 

Synopsis of talk.

Describe the history and set the agenda of systematic biology.

Classification and phylogeny both rely on character analysis (explicit or implicit)

Phylogeny an approach to reproducible methods of classification, later developed into tool for interpretation.

 

1

Why classify ?

 

The archetypal, lay view of a biologist is that of a 'systematist' Classification serves:

1) as a tool for communication

2) book keeping (library)

3) inspiration to gather information and test assumptions ('Swedish Glob' = bivalve, no shells).

 

 

Example:

Duck, crow, herring. The necessity of total information. Partial information -> classification impossible.

 

2

Archaic classification

Definition of language & the philosophical background (serves communication, organisation)

Characters used to associate with a concept.

The duck has a beak, wings, no gills and says 'quack' = it is a duck.

 

 

Discuss example from start.

There is no right answer to this!

Classify figures as 'smooth' and 'angular; 'striped' and 'dotted', 'small' and 'large'.

 

Evaluate points 1-5:

1) We use characters (properties) to classify the objects.

'Chair': 3-4 legs and some sort of seat; 'chicken': two legs, beak, feathers, source of eggs and meat; utilitarian definitions for communication.

2) We selectively apply some and ignore other properties. Subordination; membership of class defined by characters, -> classification has restricted scope.

3) We allow variation even within characters. In example: 'large' vs. 'small'; 'smooth' vs. 'angular'. Wings of duck, crow, chicken, ostrich, dodo

4) Classification may or may not be hierarchic. 'classes of classes', individual, species, genus, family, etc.

5) Classification is separate from taxonomy (giving names). Classification is not taxonomy! The example classified figures without giving them names.

3

Classic classifications - Aristotle

Plato (429-348 B.C.), Aristotle (384-322 B.C.), 'essentialism'.

Discuss 'artificial' vs. 'natural' classifications.

'Artificial': organisms fixed, sharp boundaries between groups, often based one or few characters.

'Natural': organisms vary, infinite intermediates, often information overflow, 'fuzzy' groups. You may share eight of ten characters, membership by majority rule, but not an absolute criterion.

 

Subordination of characters: Aristotle basically uses blood and eggs.

Qualitative assessment: each level is a poorer version of the level above.

System is polar - there is a 'high' and a 'low' end of a linear scale. Humans and gods are on top, justification of social strata - political and philosophical basis/interaction.

We recognize groups.

 

4

Classic classifications - the rigid system (Ray & Linnæus).

 

Until the 17th century, large European organisms (no invertebrates, fungus, nothing exotic).

 

John Ray (1627-1705) - religious agenda (ideology of the day); organizing everything known; classification is a tool of communication.

 

Carl Linnæus (1707-1778) was the first to classify all known organisms, European and exotic, large and small, plants and animals. - religious agenda (ideology of the day); organizing everything known; classification is a tool of communication. Used sexual characters. Goethe (1749-1832) wanted botanical text books censored!

 

Both: species real, invariable; they were created constant and separate.

 

Linnean system - hierarchic organisation, used today, incorporates all known organisms, invertebrate groups, exotic species, rigid hierarchy = practical and compact, lacks embellishments. Linnæus described a predefined, invariable system in organised fashion, applying a limited set of selected characters, creating an 'artificial' classification.

 

5

Clash of dynamic and rigid systems (Cuvier & Lamarck)

 

Georges Cuvier (1769-1832): All organisms invariable results of creation, 'catastrophist', acknowledged fossils as remains of living organisms, fossil organisms unrelated (they were created at different times), but built by the same structural plan as extant. 'artificial' system, subordination of characters. Brilliant anatomist and relied far more on internal anatomy than for example Linnæus. Constructed a branching classification based on anatomical characters with four major branches:

 

Jean Baptiste Antoine de Monet, Chevalier de Lamarck (1744-1829): All organisms variable results (steps of a process) of a long gradual evolution; keeper of insects, shells and worms at the Musée de Histoire Naturelle, made first keyed field guide to French flowers, subdivided the invertebrates in ten groups. Extinct species would leave a hole in the scala natura, and block the evolution; fossils remains of old specimens of living species (concealed in Alaska, Arafura Sea, Amazonas, etc.).

 

6

Lamarckism

 

Gregor Mendel's work forgotten until around 1900, comprehension of inheritance nebulous. Charles Darwin had no clue about inheritance.

Darwin's 'Lamarckism' was a better explanation than nothing.

 

What did we learn from Cuvier & Lamarck ?

 

7

Evolutionary systematists - the necessity of a fuzzy picture

 

Evolutionary thought had little impact on biological classification.

Linnean system [strict nested hierarchy and putatively invariable categories, defined by one or a few characters] still used; we changed our minds about the mechanisms, but the system survives.

 

Post-Darwin adopted Lamarck's principles, the fuzzy definition of categories and infatuation with 'intermediates'.

In 1866 Ernst Haeckel (1834-1919) presents a tree, representing evolutionary history. He coined the term 'phylogeny' to describe an inferred evolutionary pattern and established the principle, that classifications should reflect evolutionary history.

 

Johannes Thiele (mollusc systematist around 1890-1920): 'Archaeogastropoda' (the ancient gastropods), 'Mesogastropoda' (the intermediate gastropods) and 'Neogastropoda' (the new gastropods). Intermediate forms are the evidence of evolution. Philosophically different from Linnæus & Lamarck, technically the classification is not

 

8

Modern classifications (phenetics, cladistics)

 

Phenetics: classification, philosophical basis of cladistics; objective to create distinct, clearly delineated groups

 

Cladistics, one technique for modern phylogenetic analysis; describe events and classify results of those events; reflect evolutionary history; chart the evolutionary history and use it to interpret biological patterns.

 

9

Summary of classifications

 

10

Phylogeny (app. 1960s onwards)

Definition: hypothetical evolutionary history of organisms.

Techniques to construct phylogenies from phenetics.

Biology vs. palaeontology.

Phylogeny is reproducible methods of classification, later developed to be interpretation tool; classification is not the ultimate goal, it should be a tool for interpretation.

 

Summary of hand-out glossary.

Note list of literature for future reference - don't read now, but use as an introduction later.

 

11, 12

Classic examples

 

The rachidian is reduced in Mumbo jumboensis.

The diverse habitats of the Faraway Islands fostered a rapid speciation of Mumbidae.

A reduced osphradium in Mumbo jumboensis is an adaptation to ...

Groups with crawl-away larvae speciate more readily than those with planctonic larvae, because of limited dispersal.

The horse has a primitive stomach.

 

Most horrible! Postulate processes, infer ancestors and polarity of evolution. These are profound results, not just cursory remarks.

13

Establish the chronicle.

 

Task of phylogeny: Estimate evolutionary chronicle by reproducible techniques applied to explicit data. Phrasing explicit hypotheses, testing hypotheses. Historians and archeologists do this better than biologists.

 

The 'Columbus story'.

Present evolutionary narrative - "telling the tree" - keep narrative and chronicle apart. Infer an historic pattern and subsequently interpret it.

 

14

Claus Nielsen example

 

Good phylogeny, questionable technique.

Phylogeny of animals and molluscs. Note nacre as primitive molluscan character.

15

Phylogeny Techniques

Phylogeny is a concept, not plea for specific technique; use reproducible methods.

Cladistics (one technique) = character analysis applying parsimonious interpretation of anatomy, morphology, DNA, any discrete characters.

Assumptions

Can infer/reconstruct evolutionary history from living organisms ?

Parsimony

Decision method for ambiguous data, not assumption of evolution

Evolution not parsimonious

Not all characters optimized!

Taxon sampling

Characters

Independent, more characters than taxa.

Anatomy, DNA, behaviour

Homology, homoplasy (parallel evolution)

Synapomorphies provide information (resolution), autapomorphies and symplesiomorphies do not.

 

 

Theoretical example

 

Construct tree from simple A,B,C,D matrix

Some characters give information, some don't (minimum to individuals must have each alternative state - e.g., 0 and 1 - for it t carry information of common origin). We already know A,B,C,D are different.

Characters may be ambiguous

 

16

Character analysis & practical example

1) Make character diagnosis, describe as discrete states (0,1,2,...) (17)

2) Construct matrix of observed characters x studied organisms (18).

3) Run analysis on computer (PAUP or other program) & get tree (19)

17, 18, 19

Other methods

 

Maximum likelihood

biologists: DNA

DNA, total evidence

Stratophenetics

shape through time

Palaeontology: most organisms that ever lived are dead

 

 

Phylogeny as an interpretation tool

Evolutionary history

Trees are hypothesis, not tests!

The tree is a chronicle, use it to construct narrative

 

19

Phylogeny as an interpretation tool

Character evolution

Nacre (mother-of-pearl) is ancestral/primitive in mollusks.

Construct tree (20)

Map character on tree (21), assuming ancestral state.

Count number of evolutionary events (when character evolves or is reduced) given hypothesis of ancestral state (nacre evolves 1 time, is reduced app. 20 times)

Try alternative hypothesis (nacre evolved several times, OH22)

Count number of evolutionary events, given alternative hypothesis (app. 8 events).

Re-examine character's (nacre's) crystallographic properties (23) - there are four different kinds of nacre. Same number of events, but better explanation - four different structures evolved independently.

Phylogeny tells us, we have a problem with the character evolution; the phylogeny is the evidence the classic hypothesis is false. The crystallographic analysis helps us to resolve what happened: the four kinds of nacre look similar, but are different. This is Willi Hennig's 'reciprocal illumination'

15, 20, 21, 22, 23

Phylogeny as an interpretation tool

Biogeography, environment

Idealized A,B,C,D,E,F,G,H,I,J,K,L example

Invasion from mainland to island.

One invasion followed by multiple speciations.

One invasion from river to lakes.

Very idealized! Nature never that nice!

24

Phylogeny as an interpretation tool

Biogeography, gradualism of Lepechinia

Invasion from highland to lowland.

One invasion followed by multiple speciations.

NOT gradual adaptation to dryer climate.

Congruence between tree and geography.

25