Abstract/Summary

Changes in the rate at which biological evolution proceeds are widespread and common. Advancements in methodology make it possible to characterize such historical evolution more accurately and reveal complex scenarios where rates vary among organismic groups and even between individual lineages. The work presented here embraces such complexity, seeking to exploit phenotypic rate variation to reconstruct patterns and processes of evolution deep in time with unprecedented accuracy. Chapters 1 and 2 demonstrate for the first time that it is possible to reveal historical, directional trends in morphology that played out over millions and millions of years using only data from living species. These trends arose as a consequence of rapid and repeated instances of directional evolutionary change and the approach employed to detect them may be the only way to study historical adaptive trends in morphology that cannot otherwise be observed in the fossil record. Where evolution is fastest, natural selection has acted more strongly; this idea is developed further in Chapter 3 which presents a novel way to characterize an exceptional subclass of rates of morphological evolution that can be defined as positive phenotypic selection. In both Chapters 3 and 4 it is shown that such intense episodes of natural selection have punctuated the evolution of diverse groups including plants, dinosaurs and hominins. Chapter 5 demonstrates that it is possible to uncover explicit underlying causes of positive phenotypic selection and takes us one step closer to being able to truly understand the drivers of natural diversity. As a complete work, this thesis harnesses and exploits phenotypic rate heterogeneity to inform inferences about patterns and processes of evolution deep in time and to understand how natural selection has acted to sculpt morphology, giving rise to the diversity we observe today both in living species and the fossil record.