GEOL2301: Palaeoecosystems

• Do Vermin, Vermes and Vertebrates make 'good' groups for classification?

• We'll come on to a partial answer at the end of this session.

  What's important is to think about the kind of things that might unite each grouping, and whether these classes of features are the sorts of things that we might want to base a classification system upon.

Natural groups

A phylogeny is the evolutionary history of a lineage, most clearly represented as a phylogenetic tree. A tree implies a number of branching points (nodes), corresponding to cladogenesis: one lineage becoming two. Between each pair of branching points is a branch, corresponding to some period of evolutionary time, in which a lineage may have accumulated morphological change.

Morphological changes that arise once (i.e. on a single branch) and are retained in all descendants are the backbone of phylogenetic reconstruction. Such characters define a monophyletic group (from greek mono-, single, -phylum, tribe; also referred to as a clade), which comprises a single common ancestor and all its descendants. A paraphyletic group contains a common ancestor and some, but not all, of its descendants, whereas a polyphyletic group (many-tribes) comprises lots of different groups, but not their common ancestor.

Test your understanding

• What sort of group is: fish?
• Monophyletic
• Paraphyletic
• Polyphyletic
• Paraphyletic: The last common ancestor of all fish was itself a fish, but not everything descended from this common ancestor is itself a fish (e.g. tetrapods)

• What sort of group is: 'flying animals'?
• Monophyletic
• Paraphyletic
• Polyphyletic
• Polyphyletic: Birds, bats, bees and butterflies are four distinct groups; their last common ancestor was surely simple, and certainly unable to fly.

• What sort of group is: birds?
• Monophyletic
• Paraphyletic
• Polyphyletic
• Monophyletic: The last common ancestor of all birds was itself a bird, and so is everything descended from this common ancestor.

• What sort of group is: life?
• Monophyletic
• Paraphyletic
• Polyphyletic

• All life on Earth shares a number of characteristics – such as the letters of its DNA code – that indicate that everything that lives today shares a common ancestor. Everything descended from that ancestor is part of 'life', which would make 'Life on Earth' monophyletic.

  But: could life exist elsewhere in the universe? If so, it (almost certainly?) arose independently, which would make 'life' polyphyletic.

• What sort of group is: worms?
• Monophyletic
• Paraphyletic
• Polyphyletic

• The last common ancestor of all extant worms was itself a worm, but not all of its descendants belong to the group 'worms'. This is the definition of paraphyletic.

  But: arguably, not all worms are 'wormy' due to common ancestry. Mollusc worms, slowworms and others can be shown to have arisen through secondary simplification of a decidedly different body plan. One could argue that these examples of 'worminess' should be considered polyphyletic – i.e. separate innovations of 'worminess'.

• What sort of group is: dinosaurs?
• Monophyletic
• Paraphyletic
• Polyphyletic

• The last common ancestor of all dinosaurs was itself a dinosaur. But were all its descendants dinosaurs? Under the conventional definition, no: birds are descended from a paraphyletic dinosaur 'grade'.

  But one could argue that the concept of dinosaurs should be extended to include birds – which would make dinosaurs (including birds) a monophyletic group.

Tree thinking

If you're not familiar with reading phylogenetic trees, it's worth becoming so.

Baum et al.'s 'Tree thinking challenge' is a handy primer.; Forey's Introduction to cladistics goes a little deeper.

In short, relatedness should be calculated by looking for a most recent common ancestor. Everything descended from an ancestor is more closely related to each other than to anything that doesn't share that ancestor.

• On the basis of the tree on the left, is the frog more closely related to the fish or the human? Does the tree on the right change your mind?
• The fish
• The human

• Both trees depict identical relationships. The reflects common ancestors. Node x depicts the most recent common ancestor of frogs and humans; this ancestor is itself descended from the common ancestor of frogs and fish (node y).

  This is like having a grandmother ('x') in common with your cousins, to whom you are more closely related than your second cousins (who only have a great-grandparent, 'y', in common).

Fossils, particularly old ones, are often harder than living taxa to classify. This is because, on account of their ‘primitive’ condition, they often lack some or all of the key characters that identify a modern group. (Cladogenesis is not necessarily accompanied by immediate morphological change.) Rigid Linnean terms – the hierarchy of kingdom, phylum, class, order, family, genus and species – lack the flexibility to accommodate early offshoots of the evolutionary lineages leading to extant clades.

The stem group concept is what allows ‘transitional forms’ at the ‘fuzzy edges’ of categories that are defined based on living taxa to be related to a rigid hierarchical system.

• Classify the fossil taxon represented by the green flag [ ] in the figure. What characters support this classification?
• Stem-group primate
• Stem-group (rodents + rabbits)
• Stem-group mammal
• Stem-group tetrapod ('four-legs')
• The green flag is more closely related to mammals (= primates, rodents & rabbits) than any other living group in this tree. Its most recent common ancestor with mammals is a descendant of the most common recent ancestor of tetrapods. It will share certain features (e.g. hair) with mammals, but may not yet be derived enough to exhibit all the features that make a mammal a mammal (for example, maybe it lacks lactation?)
• Classify the fossil taxon represented by the orange flag [ ] in the figure. What characters support this classification?
• Stem-group amphibian
• Stem-group mammal
• Stem-group reptile
• Stem-group tetrapod ('four-legs')
• Stem-group fish

• The taxon represented by the orange flag is more closely related to tetrapods (= amphibians, mammals & reptiles) than any other living group in this tree.

  Note that this taxon happens to lack the distinctive four-legged morphology that characterizes all extant ('crown-group') tetrapods. But affinity is determined by relatedness (common ancestry). Morphology is just a tool that helps us to reconstruct relationships.

Defining the ranks

How are the higher taxa defined? Linnaeus (1735) might almost have used the aisles of his local pet shop to define his six ‘body plans’. Cuvier (1817) emphasized the disparity of Linnaeus’s “Vermes”, instead using the pre-Darwinian concept of a “chain of being” to classify organisms by complexity, and thus sophistication – with man just one step lower than the angels.

Until the 1990s, groups were defined based on a shared evolutionary history, inferred from shared homologous features – for example, body segmentation. But morphological characteristics are prone to convergent evolution: similar features – indeed similar overall morphologies – may arise in distantly related organisms in response to similar physical, environmental or ecological pressures. In the new age of cheap molecular sequencing, some long-standing relationships have been overturned on the basis of shared genetic sequences that can only be so similar (we assume) if they were inherited from a common ancestor.

How different do things need to look before they are different phyla / classes / orders? There’s no hard and fast rules, but higher taxa seem nevertheless to be useful units of study.

Consider the pros and cons of using different taxonomic levels to gauge diversity. What sort of diversity might a rank capture? How might definitional uncertainties bias a count?

• What does the number of different phyla tell us about an assemblage?
• • Might capture very high-order, evolutionary-scale patterns
  • Many phyla might be adapted to certain environments (e.g. marine burrowing), few to others (e.g. terrestrial mountaintops), more for evolutionary than ecological reasons.
  • Phyla are defined by ignorance (about relationships), which is conceivably an artefact of extinction (a chance macroevolutionary process), rather than anything ecologically meaningful. They might tell an interesting story, but it requires careful thought to unravel the question of what, if anything, a phylum represents from grander stories of ecological and environmental change.
• What does the number of different species tell us about an assemblage?
• • Species are a useful fine-grained measure of diversity, very useful for ecological analysis.
  • We might worry about 'lumpers' and 'splitters' when comparing different assemblages from different parts of the world, as how 'broad' the definition of a species is can be subjective.
• What does the number of different genera tell us about an assemblage?
• • A genus seems like a good unit of diversity; even if we miss out on some species-level signal, species within a genus usually have a similar ecology.
  • Genera follow a Pareto distritribution, hinting that they represent meaningful entities.
  • Counting genera through time might also reveal something about grand-scale long-term patterns in evolution.

Phenetics aims to group taxa according to their similarity, rather than their evolutionary relationships. Taxa that look similar are grouped together, whether or not their similarity is inherited from a common ancestor. (For this reason, phenetic groups are often polyphyletic.) Phenetics can sometimes get a bad press, and there is no consistent framework for ‘phenetic classification’, but it offers a neat way of exploring how the nature of organisms has changed through time.

Organisms might be grouped by morphological similarity, or by ecological similarity. A fundamental ecological distinction might be made between autotrophs (self-feeding – often photosynthetic) and heterotrophs (other-feeding – grazers or predators, perhaps). A more nuanced ecological classification uses three primary dimensions of tiering, feeding and motility, which can be exploded into a 6×6×6 cube; the occupancy of this cube through time offers an interesting window on how ecological possibility has been more thoroughly explored through the Phanerozoic. An equivalent endeavour is the mapping of disparity (diversity of form) through time. On a finer scale still, an ecosystem might be divided into individual niches – particular living conditions – with each niche occupied by a distinct species that is adapted for that particular niche. Communities that look very different – perhaps separated by space or time – may have a similar underlying ecological structure; equivalent niches may have different occupants yet the overall structure of the ecosystem may be the same.

• Why might large-scale trends in ecological and taxonomic diversity be decoupled? Would you expect to see the same pattern at lower taxonomic / ecological ranks?
• • Taxa are mainly defined by extinction, which is backward-looking, and has applied throughout the tree of life over the past 500 million years. Many of the gaps between phyla can be traced back to the Cambrian period, when taxonomic diversification happened quickly.
  • Ecological diversity requires a conducive environment to be supported.
  • Did it take a while for the ocean to be oxygenated, and sufficiently productive, to support energy- and oxygen-intensive modes of life, such as fast, deep-burrowing mining?
  • Could this trend towards productivity itself be biologically mediated, with gradual ecological escalation driven by increased efficiency of primary production supporting larger populations, increasing energy availability and thus permitting ever more active modes of predation, leading to ever more active prey?

Here are some optional long-form questions to help cement your understanding of this week's content.

• What are the advantages and disadvantages to the three main approaches to classification?
• How might our perpsectives on diversity differ under (i) a Linnean, rank-based approach to classification; (ii) a cladistic, stem-and-crown approach to classification?

• Headlines: A Linnean approach might lump together different 'body plans' (e.g. birds & dinosaurs, molluscan classes), making a rank look like it contains more or less diversity than it really does. Counting ranks could oversimplify the true diversity / disparity.

  A cladistic approach allows a more precise (and meaningful?) classification of fossil material, though is not necesssarily as amenable to a quantitative approach. As the transition from stem to crown group is simply defined by what happens not to have gone extinct, it perhaps makes more explicit the basis on which group membership is defined.