Weighing the Evidence for Evolution

Evaluate asks you to make a judgement based on criteria, determining value or significance. You need the strengths, the weaknesses, and a judgement that follows from weighing them, not a list with a verdict bolted on.

5 marks. Say what biogeography is and what it predicts, give the strongest evidence for it, give the genuine weaknesses, then judge, ideally by saying what it does that other lines cannot.

Biogeography is the study of how species are distributed across the Earth. If species arose by descent with modification from ancestors in particular places, then distribution should follow geography and history: related species should tend to be found near one another or on landmasses that were once connected, and isolated areas should hold distinctive species found nowhere else.

The predictions are met in detail. Australia's long isolation is accompanied by a distinctive mammal fauna dominated by marsupials and monotremes and by very high numbers of species found nowhere else, which is what descent in isolation predicts. Related groups are distributed across landmasses that were formerly joined as Gondwana, matching the geological reconstruction rather than matching present-day climates. Island faunas characteristically resemble those of the nearest mainland while differing in detail, which is what colonisation followed by divergence predicts. Oceanic islands also characteristically lack whole groups such as large terrestrial mammals, which is what a dispersal requirement predicts and which special creation in place does not.

The line has real limitations. Distribution is the product of evolution, geology and chance dispersal together, so untangling them is difficult and some cases remain disputed: a group present on two separated landmasses may reflect an ancestor split by continental drift or a later chance crossing, and the two explanations can be hard to distinguish. Distributions are also erased and overwritten by extinction, glaciation and, recently and on a large scale, by human introductions, so the present pattern is an incomplete record of the historical one. And biogeography is evidence for common descent and for the geographic mechanism of speciation, but it says relatively little directly about natural selection as the mechanism of adaptation, which is the second half of what the theory claims.

My judgement is that biogeography is strong evidence for common descent and for the role of isolation in generating diversity, and weak evidence about natural selection specifically. Its distinctive value is that it is independent of anatomy and molecules, drawing instead on geology and geography, so when it agrees with phylogenies built from morphology and from DNA the agreement is between genuinely unrelated methods. That independence is worth more than any additional anatomical comparison would be. Its weakness is consequently not a reason to discount it but a reason to use it alongside the other three lines, each of which is strong where it is weak.

Explain means giving the mechanism, and the mechanism here is methodological: what makes one kind of evidence capable of testing another rather than merely adding to it.

5 marks. State the condition for a genuine test, set out the competing anatomical readings, say what the molecular data predict under each, give the result, then state what has been demonstrated.

One line of evidence tests another only if it is independent of it and if the two could have disagreed. Molecular data qualify on both counts: DNA and protein sequences are collected and compared without reference to body form, and there is no mechanism by which a researcher's anatomical expectation determines what the sequence says. A result that agrees therefore carries weight, because a result that disagreed was genuinely possible.

The thylacine presents exactly such a case, because its anatomy supports two incompatible readings. Its overall body and skull form closely resemble those of a wolf, which taken at face value suggests a close relationship to placental canids. Its reproductive anatomy, including a pouch, and the detail of its skull and dentition instead place it among the Australian dasyurid marsupials, alongside quolls and the Tasmanian devil. Anatomy alone therefore gives two answers depending on which features are weighted.

The two readings make different, checkable predictions. If the thylacine were closely related to canids, its DNA and protein sequences should be more similar to those of a wolf than to those of a quoll or devil. If it is a dasyurid marsupial, its sequences should be much more similar to those of quolls and devils, and no more similar to a wolf's than any other marsupial's is. These predictions are mutually exclusive, so the data must favour one.

Sequence data, including that recovered from museum specimens of the extinct thylacine, place it firmly within the marsupials and among the dasyurids, close to quolls and the Tasmanian devil, and not close to canids. The anatomical reading based on reproductive and skull detail is therefore supported and the reading based on overall resemblance is rejected. The wolf-like form is convergent, an independently evolved solution to the same predatory role, and the similar structures are analogous rather than homologous.

The case demonstrates two things. First, a relationship proposed from anatomy is a hypothesis that independent data can confirm or refute, which is what makes classification empirical rather than a matter of judgement. Second, it shows why molecular data have become the arbiter where anatomy is ambiguous: overall resemblance can be produced by convergence, whereas sequence similarity across many positions is far harder to produce without shared ancestry. Agreement between independent methods is what turns a plausible classification into a well-supported one, which is precisely why the four lines of evidence are weighed together rather than ranked.