Search

Hominins (the clade including modern humans and their fossil ancestors) were a taxonomically and morphologically diverse group during the Plio-Pleistocene, and their evolution documents the only known transition to obligate bipedalism in primates. However,many aspects of their shared evolutionary history remain frustratingly unclear due to uncertainty about whether change in the fossil record reflects genuine evolutionary change or variation in our sampling of the rock record. Here, a comprehensive assessment of the quality of the early African hominin fossil record is presented. A specimen database of all early African hominin fossils (>5000) has been compiled including taxonomic, geological, anatomical, and bibliographic information. Using a range of sampling metrics (fossil-bearing formations, collection effort, sampled area, and ghost lineage diversity), it is shown that the pulsedlike pattern of uncorrected (taxic) hominin diversity is almost entirely controlled by rock availability. By contrasting taxic with phylogenetically corrected diversity, hominin diversification appears unconstrained through the lateMiocene and Pliocene, with diversity constantly increasing until a single peak is reached in the early Pleistocene. Phylogenetically corrected diversity shows no discernible link with sampling metrics and there is no direct evidence that shifts in climatic conditions drove diversification. A study of specimen completeness through geological time shows that while sampling metrics (specifically sustained collection effort at rich deposits) have a major influence on patterns of specimen completeness, specimen completeness has only a moderate influence on diversity patterns. It also shows that specimen completeness is poorest during the period most pertinent to human origins, the estimated Pan-Homo divergence date, in large part due to under-sampling (< 4% of Africa by sampled area). In combination, this work illustrates that the hominin fossil record is by no means an unbiased depiction of evolutionary events, and therefore its quality and incompleteness should be fully understood before any interpretation of macroevolutionary patterns.

This thesis describes the development of an assay for free radical scavenging (antioxidant) activity in biological fluids based on an enhanced chemiluminescent reaction. Light emission from the reaction depends on the constant production of free radical intermediates and is therefore sensitive to interference by antioxidant compounds. The time period of light suppression is directly related to the amount of antioxidant added to the reaction. In this way the antioxidant activity of biological samples can be related to a standard antioxidant solution of the vitamin E analogue, trolox. The effect of a variety of pure compounds upon light emiSSion are described. Based on these observations a mathematical model for the reaction kinetics in response to the addition of antioxidants and other compounds is developed. The impact of more complex biological fluids on the reaction is described with particular reference to serum. An assessment of the contribution of individual antioxidants to serum total antioxidant activity suggests that urate accounts for 70% while ascorbate and vitamin E each account for a further 10%. The use of the assay is extended to the measurement of antioxidant activity in solutions of plasma lipoproteins isolated by density gradient ultracentrifugation. These studies suggest that vitamin E is the major (but not exclusive) contributor to lipoprotein antioxidant activity. The distribution of antioxidant activity across different lipoprotein fractions and co-operative interaction of antioxidant activity in lipoproteins are described.