Biology Reference
In-Depth Information
be explained, without attempting to predict from an ESS model the sexual
dimorphism in a particular species.
(iii)
In some cases, we should not necessarily hold quantitative predictions as the
Holy Grail, because the development and testing of qualitative models will be
the more powerful approach (Frank, 1998). ESS models will always be
oversimplifications of the real world. Furthermore, it may be possible to develop
alternative theories that make the same quantitative prediction. In contrast,
qualitative predictions about how traits should vary in response to variation in
key parameters can usually be made much more unambiguously. For example,
in Chapter 10, we discussed Hamilton's local mate competition (LMC) theory,
which provides one of the greatest success stories in the field of behavioural
ecology, having been supported in a huge range of taxa, from malaria parasites,
to worms, to insects, to mites, to snakes. The majority of the empirical support for
this theory has come from testing the qualitative prediction that females should
lay more female-biased sex ratios when fewer females lay eggs on a patch (Fig.
10.4), and not from testing whether it quantitatively predicts average population
sex ratios. This makes sense from a theoretical perspective - a slew of models
have investigated the consequence of incorporating various life history details
into Hamilton's model, with the general result being that, whilst they alter the
quantitative predictions, the qualitative prediction is robust (West, 2009).
To give another example, Darwin's (1859)
Origin of Species
had little
quantification. What made his argument so compelling, nevertheless, was that
many independent lines of qualitative evidence (from population growth,
geographical distributions, variation, embryology and so on) all pointed to the
power of evolution by natural selection.
It is possible to carry on discussing the pros and cons of optimality models for a long
time, but the strongest argument in their favour is that, over and over again, optimality
arguments have helped us to understand adaptations. Although alternative approaches
have been suggested (Gould & Lewontin, 1979), they just haven't proven very useful. We
have illustrated the use of the optimality approach in the preceding chapters with
behavioural examples - foraging, flock size, territory size and so on - but optimality
arguments can equally well be used to understand adaptations at the physiological and
biochemical level. For example, the familiar 'herring bone' arrangement of the swimming
muscles of many fish is not merely an incidental design feature. This arrangement allows
the muscles to contract at a rate which maximizes their power output (Alexander, 1975).
At the biochemical level the energy for muscle contraction is generated by oxidation of
carbohydrates or fats via the Krebs cycle. It would be chemically feasible to carry out the
oxidation by a more direct route, but the advantage of the cycle is that it maximizes the
net energy gain per molecule oxidized (Baldwin & Krebs, 1981).
Causal and functional explanations
Behavioural ecology is about functional explanations (the answers to 'why?'
questions) of behaviour. As emphasized in Chapter 1, a great deal of misunderstanding
can arise if functional and causal ('how?') explanations are confused.
