The Evolution of Social Organisms: Modelling Reproduction Strategy

The focus of this thesis is on the study of reproduction strategy in the context of evolutionary and social-evolutionary theory. Much of the hierarchical structure that is evident in the natural world is due to major evolutionary transitions where individual subunits that once reproduced individually now reproduce only as part of a larger unit. Modelling and understanding the processes behind the evolution of this hierarchy can have applications in both biology and computer science. I argue that to explain the major transitions it is necessary to understand why an individual would reduce its reproductive success to invest instead in a higher reproductive process (i.e., reproduce collectively with other individuals). To address this problem, a method for studying reproduction strategy was developed and is presented in this thesis. The method takes an abstract physiological approach to reproduction. It considers an individual as a quantity of resources and set of genes which define its reproduction strategy. I then investigate the advantages of different reproduction strategies and identify which strategies may dominate others. The strategies considered in my investigations include: an individual reproducing on its own; an individual gambling its total resources against those of multiple other individuals; or an individual sharing its reproductive effort with a partner or several other partners. Starting with individual reproduction, I simply study why an individual might reduce its reproductive rate when, on the face of it, it seems that maximum fecundity should be the best option. The model is also motivated in light of current literature on life history and microbial ecology in particular. The results show how it can be advantageous for an individual to hedge its bets and delay reproduction; waiting instead until it has accumulated more resources and is less vulnerable to harsh periods. The results make predictions that are experimentally verifiable. Given the model of individual reproduction, I reapply the growth equations to question whether there is any advantage to sharing reproductive effort through collective reproduction. This model also shows that it can pay to hedge one’s bets and invest in the less vulnerable, but slower, collective reproductive strategy. The results show that there is a mathematical relationship between the number of parents and the up-front cost of reproduction spent on creating a new offspring – depending on the extra cost per parent, two parents may be the best strategy or perhaps many parents. Looking in more detail at the transition from unicellular organisms to multicellular organisms, I model the macrocyst stage in the slime mould Dictyostelium. I consider how the macrocyst stage may be an early example of collective reproduction in protozoa. Here individuals aggregate to be ingested by a central cell which produces homogeneous offspring. I assume that each individual is gambling on being the central cell and the model presented reveals under what conditions this is likely to be a good strategy when compared to individual reproduction. Again, the results show that there is an advantage to hedging one’s bets and investing in the macrocyst rather than going it alone. Finally I consider the origin of sexual reproduction in more detail. The traditional approach argues that the slower growth rate of sexually reproducing organisms means that there is a paradox concerning the origins and maintenance of sexual reproduction, especially when one considers males which do not contribute to their offspring. Extending the previous model of collective reproduction, I consider how many resources selfish individuals may contribute to their offspring. The results show that there is a lower bound to the resources individuals may contribute and that when there is a high amplitude of resource fluctuation, the sexual strategy can dominate an asexual strategy. As well as the main body of work on the topic of individual reproduction, some further background work is also presented. The models use both mathematical and computer simulation models. These two approaches are compared and contrasted with reference to their value in generating good scientific explanations of the sorts of phenomena found in the types of systems I am studying.