UC Berkeley

With more than 260,000 species, the angiosperms are the most diverse group of land plants on earth today. Many would argue that their striking diversity stems from the acquisition of the flower along this evolutionary lineage. The argument goes that by enclosing the plant's sex organs, especially the ovule, the flower provided angiosperms with special means to withstand a wide range of environmental conditions, while facilitating pollination or pollinator attraction and seed protection and dispersal. Regardless, the diversity of shapes, colors, and sizes of flowers across the angiosperms is irrefutable and fascinating. Understanding the mechanisms that underlie flower diversity leads us to the understanding, at least in part, of how evolutionary processes have enabled the origin of different forms in nature.

Although the Modern Synthesis has provided a solid framework for understanding how genes evolve in populations, it lacks a theory to satisfactorily explain the evolution of morphological diversity, as it largely marginalized the role of development in the evolution of biological form. Recently, however, an increasing attempt to understand the interrelationships between evolution and development has emerged as a new research field known as evolutionary developmental biology, or, for short, evo-devo. The study of genes involved in different developmental processes, and how changes in these genes or on their regulation can lead to changes in organismal form has become an insightful field.

This dissertation focuses on the evolution and diversification of floral morphology in the Zingiberales and their implications for our understanding of the evolution of plant bauplan. The tropical monocot order Zingiberales provides an excellent framework for evolutionary developmental studies, as changes in floral form throughout the evolution of this group are mainly due to changes of form and function in the petal and stamen whorls, where stamens become infertile and petaloid.

The first part of this dissertation describes how changes in classical floral organ identity genes result in changes in floral organogenesis throughout the evolution of the Zingiberales. First, through a combination of careful morphological studies and genetic approaches, I establish the homology of floral organs, particularly the nature of the so-called `petaloid appendages' on fertile stamens of the ginger group. Second, I show that positive selection is acting upon the AGAMOUS (AG) lineage, and changes in the AG protein suggest a mechanism capable of explaining the morphological changes observed in the Zingiberales flowers.

The latter part of this dissertation goes beyond organ identity genes to investigate the development of organ morphology. In this section, I demonstrate the involvement of the abaxial-adaxial (ab-ad) polarity gene network on the evolution of filament morphology, not only within the Zingiberales but also across all angiosperms, and provide evidence that morphogenetic processes, not just organ identity per se, are driving the evolution of floral form across the order. By studying ab-ad polarity genes, well-known for the establishment of abaxial and adaxial surfaces of leaves, sepals, and petals, I show how the same gene regulatory network has been co-opted during the evolution of angiosperms to shape filament morphology in flowering plants.

I conclude this dissertation by discussing the implications of these findings to our understanding of the mechanisms of plant bauplan evolution. Lastly, I analyze the floral evo-devo research program through a historical and philosophical perspective, hoping to shed light on future directions of research in the field of plant evo-devo, as a consequence of important conceptual changes that this field has undergone in the past two decades.