This thesis presents findings that suggest that the survival of the fittest is synonymous with the survival of the stable. Richard Dawkins laid out a compelling framework for looking evolution from the perspective of the gene. The Gaia hypothesis and niche construction point to the coevolution of gene and environment. Under the assumption that gene and environment co-evolve it is found that there is strong tendency for a system to seek stable solutions which may or may not optimise the organisms to the environment. It is argued that this thought should be the guiding principle when looking at the evolution of natural and man made systems. The thesis uses the framework of the Daisyworld model developed by Watson and Lovelock to demonstrate these ideas. To this end the original Daisyworld model is analysed and an analytical solution for the equations is provided. The analyses suggest a maximisation of growth rates in an system with evolving albedos. This behaviour is then compared to an two dimensional model of the Daisyworld and the differences are clearly enumerated. Following this an extension is made to the two dimensional model to allow the evolution of mutation rate. The model is then shown not to maximise growth rate, but instead stability. It also shows the evolution of levels of selection higher than the individual or gene and suggests that this is completely compatible with a gene level perspective. It is thus argued that the focus of evolution is not individuals or genes, but rather the systems of which they are a part.

This thesis argues that Samuel Butler's evolutionary work represented one of the most eccentric and enlightening alternatives to 'natural selection'. In his scientific research, the English writer used 'memory' as a justification of the process of heredity. This represented, at the turn of the twentieth-century, a shift from the mechanical approach to natural selection to an idea of evolution based on a psychophysiological process. First, my work addresses the methodological difficulties that occur in placing Butler's work within the late Victorian debate. Neither the Victorians themselves, nor subsequent scholars, have known what to make of Butler, most usually attempting to place him in the category of literature whilst ignoring the substance of his scientific work. Looking at the reception of Butler's ideas in England and Europe, and critically discussing the problem of Butler's identity in the context of contemporary scholarship, I create a new picture of Butler as a Victorian amateur 'scientist'. Second, this research discusses Butler's popularisation of science, looking at his use of language, style and form in his fictional and non-fictional works. It also discusses Butler's work of translation and his popularisation of European scientific hypotheses within the British 'marketplace' of science. Third and finally, this work explores Butler's hypothesis of unconscious memory in relation to similar cases in Europe. Looking at the French and German psychophysiological debate, it places Butler's evolutionary work in a pan-European context. Overall, this thesis attempts to create a three-dimensional intellectual image of Butler as writer and participant in science, and in doing so finds that Butler is a problem case who tests the edges of scholarship in history and literature of nineteenth-century science.

The Ecdysozoa is a large clade of animals comprising the vast majority of living species and some of the most studied invertebrate models, including fruitflies and nematodes. Some of the relationships between major ecdysozoan groups remain uncertain, however, undermining comparative studies and impairing our understanding of their evolution. One hotly debated problem is the position of myriapods which have been recently grouped according to molecules with chelicerates and not with insects and crustaceans as predicted by morphological evidence. Other disputed problems are the position of tardigrades, the position of hexapods within the crustaceans as well as the mutual affinities of the nematodes and priapulid worms. Molecular systematics of the ecdysozoans is complicated by rapid divergence of the main lineages (possibly evidenced in the Cambrian explosion) followed by a subsequent long period of evolution. This may have resulted in a dilution of the historical phylogenetic signal and an increased likelihood of encountering systematic errors of tree reconstruction. This problem is exacerbated by many lineages being poorly represented in current molecular datasets, as sequencing efforts have been biased toward lab models and economically relevant species. In order to overcome problems of systematic error, I have assembled various large mitochondrial and phylogenomic datasets, including new data from undersampled tardigrades, onychophorans and especially myriapods. I analysed these datasets using the most recent evolutionary models. I have developed two new models in order to describe the evolutionary processes of metazoan mitochondrial proteins more accurately. My analyses of multiple datasets suggest that the grouping of myriapods plus chelicerates found by previous authors is likely to be the result of systematic errors; I find support for a closer relationships between myriapods and a group of insects plus crustaceans (the Mandibulata hypothesis). My analyses also support a paraphyletic origin of Cycloenuralia (nematodes and priapulids) and a sister group relationships between tardigrades, onychophorans and euarthropods in accordance with a single origin of legged ecdysozoans, the Panarthropoda. Finally, results support a monophyletic group of hemimetabolan insects. The majority of the results reconcile molecules and morphology, while others shade new light onto arthropod systematics. The evolutionary implications of these systematic findings as well as methodological advances are discussed.

In this thesis a set of axioms sufficient to generate the Darwinian theory of natural selection is developed and some of the implications of the theory derived, In the process of developing the axioms the basic assumptions of the theory are clarified and the doubts raised by its apparent tautological nature resolved. The theorems derived from the axioms show that certain consequences that must be derivable in a theory of natural selection can be derived. (E.g. One theorem shows that there is always a subpopulation, fitter than the population as a whole, which is in the process of taking over the population; other theorems show that under certain circumstances less fit subpopulations will be eliminated.) They also show that the axiomatized theory has various expected consequences. (E.g. One theorem states consequences of density dependent selective advantage; several theorems show differences between natural selection in interbreeding populations and natural selection in non-interbreeding populations.) Thus the theorems provide evidence for the assertion that these axioms will generate the Darwinian theory of natural selection. Two further axioms are stated in order to indicate how the axiomatized theory of natural selection can be embedded within an axiomatized theory of evolution and further theorems are proved with the use of these axioms.

To investigate major evolutionary trends and the importance of fossil data we need to be confident that both phylogenetic trees and fossil dates are reliable. Indices of stratigraphic congruence provide a way to quantify the fit between the fossil record and phylogeny, but appear to be subject to a number of putative biases. I used both simulated data and a large sample of empirical trees to determine the effect of these biases on the most widely used indices of stratigraphic congruence to determine. The GER* (the modified gap excess ratio) was the least sensitive and therefore recommended for use. I found that stratigraphic congruence varied significantly across higher taxa (for example, arthropods were less stratigraphically congruent than tetrapods), and also throughout the Phanerozoic (the last 540 million years), closely following the taxonomic composition of my sample. I focussed on data quality and in particular taxon sampling, homoplasy and tree support to investigate general trends across taxonomic groups. A novel script was developed to automatically carry out continuous taxon jackknifing to investigate the effect of taxon sampling on the stability of phylogenetic trees. While this is a computationally intensive process, I found that measures of homoplasy and support (which are much easier to calculate) could serve as partial indicators of whether a tree was likely to be sensitive to taxon sampling. There was no major variation in taxon sampling trends across higher taxa. A modified version of this script was then used to look at particular cases of conflicting phylogenetic hypotheses to determine how easy it would be to get a data set to generate a constraint topology with only small changes to the taxon sample. In almost every case, it required maximal removal of taxa from the data set in order to match the constraint topology, indicating that there were very different phylogenetic signals in the different data sets. The extent of trends across taxonomic groups and through time is variable. Although stratigraphic congruence varies significantly between groups and throughout the Phanerozoic, measures of homoplasy and support do not appear to be taxon dependent. Taxon sampling is an important consideration when designing phylogenetic analyses: denser taxon sampling can have a positive influence on estimates of phylogenetic accuracy and perturbations of the taxon sample can result in radically different evolutionary relationships.

There are two important reasons for studying the reception of Darwinian concepts in Russia in the 19th century. First, very little is generally known in England about Russian biology and geology of that century and the period is especially interesting since it precedes that of modern Soviet science. Second, the differences and similarities of the reception of Darwinism in Russia with its reception elsewhere throw light on the general progress of acceptance and rejection of a scientific theory. Darwin's theory of evolution affected man's whole conception of himself. Since it was a philosophy as well as a scientific theory the quality of its reception in Russia as in other countries was to a certain extent dependent on the cultural mores of that country as well as on the force of scientific argument. In order to understand the reception of Darwinian concepts in Russia in its totality it was necessary to deal with the background in some detail. An outline of the cultural, social and political traditions in Russia in the 19th century provides a background against which the reception of Darwinian concepts can be seen in its perspective and in its relation with the various intellectual discussions and cultural movements taking place during that time. A detailed description of the organisation of science and of the actual content of the sciences of biology and geology in the first half of the 19th century provides the necessary historical back. ground to the state of science in Russia at the time of the dissemination of Darwin's book the Origin of Species, a background especially essential for the English reader. The actual steps by which Darwin's ideas were disseminated and the reaction from the reading public throughout the second half of the 19th century forms a skeleton in which the impact of the theory of evolution and of Darwinian concepts on Russian biology and geology and on its social and cultural ideas and of their impact in turn on the interpretation and emphases given to the theory and its concepts in Russia.

The theory of Panspermia suggests that life was brought to Earth from an external source in space. The theory can be further divided into a) Neopanspermia, the view that life continues to arrive from space to Earth, b) Pathospermia, the idea that pathogenic organisms arrive from space and c) Cometary Panspermia which specifies that extraterrestrial organisms originate from comets. This main aim of this study was providing evidence in support of Neopanspermia. Six stratospheric balloon launches were carried out in order to sample the stratosphere for microbial cells (biological entities, BE). Analysis of isolates was achieved using Scanning Electron Microscopy (SEM), Energy Dispersive X-Ray Spectroscopy (EDAX), and Molecular techniques, to help determine the biological nature of any isolates and their origin. SEM and EDAX showed that, while most of the isolated material was inorganic cosmic dust, a few were biological in nature, although generally not recognizable as known terrestrial organisms. Some of the BEs were larger than the theoretical 5 micron limit for the transfer of a particle from Earth to the sampling heights, thereby suggesting a non- terrestrial origin. This was confirmed by the lack of similar sized (i.e. exceeding 5 micron) known terrestrial organisms such as pollen, grass shards, and fungal spores. There is clearly no sieve present in the atmosphere that would allow the isolated BEs to be elevated to the stratosphere from Earth while holding back known biological forms. It was therefore concluded that the large biological entities we isolated from the stratosphere are incoming to Earth from space and continually impact the Earth. Single cell amplification and identification also showed the presence of DNA and revealed a diverse population of known microorganisms distinct from the isolated biological entities. The ability of microbes to regain viability from ancient samples was also assessed by the isolation of bacteria from amber and halite rocks, thereby providing evidence that these organisms can survive after an extended period of millions of years, a finding which is of potential relevance to the transfer of microbes in the process of panspermia. Three meteorite types were also examined for biosignatures using SEM and EDAX techniques, two showed microfossil formations which can provide evidence for the theory of cometary panspermia. Evaluation of microbial survivability in the presence of exposure to ultraviolet C when embedded in ice was evaluated, as was the potential shielding by the presence of solid inorganic particles. It was also demonstrated, that the use of visible light generated by UV from fluorescence can provide energy for Cyanobacteria, which might have been the first microbes to inhabit our planet. Showing that life can survive in such conditions without the need of a protective atmosphere made up of oxygen that is necessary to form the protective ozone layer, this may explain how the first microbes were introduced to earth in the panspermia theory.

The study of tempo and mode in evolution is rooted in palaeontology and begins with the fossil record, our only empirical record of character acquisition through geological time. With the growth of cladistic methods and increased sampling of die fossil record the amount of character information has exploded. An initial question of this study is how did clades acquire tiieir characters, in a relatively rapid early burst as lungfishes did, or by steady, continual change as synapsids did?

This study in philosophy of biology examines the conflict between the research program of evolutionary developmental biology (Evo-devo) and the traditional view of evolution, the Modem Synthesis. In the core of this conflict lies the role of development in evolutionary explanation: whereas Evo-devo uses developmental facts to explain evolutionary outcomes, according to the received view of evolution this is to misunderstand evolutionary theory. The aim of the thesis is a philosophical analysis of Evo-devo explanations and their consequences for our understanding of evolution. A number of philosophers and biologists have argued in favour of the compatibility between Evo-devo and traditional evolutionary explanations. So, it has been argued that: Evo-devo explanations aren't really explanations at all, only descriptions; they are about a different level of the evolutionary process (variation); they are explanations of macro- as opposed to microevolution; they are explanations of form rather than function; and lastly, they are individual-level rather than population-level explanations. Not all of the above distinctions point towards a peaceful co-existence of the two kinds of explanations. In order to have a more clear idea of the differences between the two, I discuss the above distinctions in light of some recent examples of Evo-devo explanations. I give an account of how we should understand the general nature of such explanations, and the difficulties in synthesizing them in a common evolutionary theory. Throughout the thesis I will discuss various important concepts and kinds of explanations used in Evo-devo: central among them are Evo-devo explanations of macroevolution, the concept of the Bauplan and similar notions like the vertebrate limb, and the notion of constraints in evolution. The central claim of the thesis is that Evo-devo shows that development can be causally relevant to evolutionary change.

Being able to determine the kinship structure and the relatedness of individuals are important prerequisite to test many models of evolution, Cepaea nemoralis is an ideal model organism for studying these processes because it has a rich history of research, and with its relatively low dispersal ability lends itself to a high degree of local adaptation over small and relatively easily studied areas. During this project we developed molecular markers to in an attempt to genotype of a population of C. nemoralis in Berrow Somerset. There are several methods available to exploit the discriminative power of DNA markers. These methods can be grouped into two main categories: moment estimators and likelihood estimators (Blouin, 2003; Thomas, 2005; Van-de-Casteele et a/., 2001). The aim of this project is to extend earlier methods of estimating relatedness and to model the probable recent pedigrees of a group of individuals purely from genetic information. This approach aims to model the multi locus genealogy back to a certain point in the past, prior to the point where the ancestral genotypes are drawn from an appropriate frequency distribution, in a similar way to (Gasbarra et al., 2006), taking into account genotypic errors. We developed a Moran model of evolution allowing for overlap of generations and different levels of mating fitness in a Bayesian framework with the use of importance sampling based on an algorithm developed by Beaumont (2003), called group importance metropolis-Hastings algorithm.