Life exhibits a remarkable diversity of patterns, yet the evolutionary processes
underlying this phenotypic diversity operate under major constraints. These include
the constraints of development which define a restricted space of readily generated
forms and patterns. There are also genetic constraints that restrict the extent to
which genotypes may diverge within and among populations. In addition there are
selective constraints that influence variation according to fitness. A major challenge
is to understand how these various constraints interact to define observed
phenotypic patterns. I will describe a range of experimental and computational
approaches to this problem using Antirrhinum (Snapdragon) as a model system.
This system has the advantage of being amenable to both developmental and
population genetic analysis. The work involves collaboration with population
geneticist Nick Barton (Vienna) and computational scientist Andrew Bangam
(Norwich).