Biology Reference
In-Depth Information
Replicating molecules
⇒
Populations of molecules in compartments
Table 15.2
The
major evolutionary
transitions, each
of which has led
to a new level of
complexity
(Maynard Smith &
Szathmary, 1995)
Independent replicators
⇒
Chromosomes
RNA as gene and enzyme
DNA & protein (genetic code)
⇒
Prokaryotes
⇒
Eukaryotes (cell nucleus & organelles)
Asexual clones
⇒
Sexual populations
Protists
Multicelled animals, plants & fungi (with cell
differentiation, i.e. organs)
⇒
Solitary individuals
⇒
Colonies (with non-reproductive worker castes)
Primate societies
Human societies (language)
⇒
Optimality models and ESSs
In nearly every chapter we have used the ideas of optimality and ESSs. Here we will
briefly recap some of the criticisms which have been levelled at optimality arguments
and some of the limitations with putting them into practice.
Criticisms of
optimality models
1
The idea that animals are optimal cannot be tested
.
As we saw in Chapter 3 this criticism is based on a mistaken notion. The aim of using an
optimality model is not to test whether animals are optimal, but to test whether the
particular optimality criterion and constraints used in the model give a good description
of the animal's behaviour.
2
It is hard to tell why the animal's behaviour does not fit the predictions exactly
.
Very often the simple models give an approximate but not exact description of the
animal's behaviour. This could be because the model makes incorrect assumptions
about constraints or goals, or because some component of cost has not been measured.
There is no simple way of distinguishing between these possibilities.
One way around this problem is to use theory to make predictions for how traits
should vary (across individuals or populations or species), rather than to predict a single
value for a trait in a population. The former relies on a qualitative prediction from
models that can be very simple and just need to capture a key aspect of the biology;
whereas the latter relies on a quantitative prediction from a possibly very complex model
that could depend upon an almost infinite number of biological details. Steve Frank
(1998) has argued that this often makes the qualitative approach the best use of theory.
3
Animals are not well enough adapted to optimize
.
The main rationale for using optimality and ESS models is the assumption that natural
selection produces well-adapted animals, the aim of the models being to find out how
they are adapted. There are, however, at least four factors which can limit the extent to
which animals are perfectly adapted.
Reasons for lack
of adaptation …
