Euglossine-orchid interactions.
Male euglossine bees (~200 spp.) collect fragrances from orchid flowers and store them in specialized pockets located in their hind legs. In so doing, males pollinate ~700 species of orchid flowers. Later during courtship, hovering males present their chemical bouquets to females. Orchids—unlike most flowering plants—do not produce loose pollen grains, but instead package pollen in compact masses called pollinaria, which are the units vectored by male bees. Different species of orchids attach pollinaria onto different parts of the body of male bees, and attract specific pollinator assemblages by producing unique chemical combinations in their fragrances.

To elucidate the possible patterns of coevolution between euglossine bees and their orchid hosts, I assembled a species-level phylogeny of the entire tribe Euglossini, and sequenced DNA from orchid pollinaria recovered from male bees caught in the field. I am also performing chemical analyses (GC-MS) of the fragrances collected by male bees to investigate three related questions: (1) the evolutionary changes in the chemical bouquets of male orchid bees; (2) the utility of male-gathered fragrances as informative characters for phylogenetic inference; and (3) whether chemical differentiation is stronger or more frequent between sympatric sister taxa than between allopatric sister taxa.

Fossil evidence of orchid pollination.
Recently, my colleagues and I described a diagnosable fossil orchid (Meliorchis caribea) from a pollinarium attached to a stingless bee preserved in Dominican Amber, 15-20 Myr old. This discovery not only constituted the first unambiguous fossil Orchidaceae, but also an unprecedented observation of a plant-pollinator interaction in the fossil record. Cladistic methods applied to a morphological character matrix resolved the phylogenetic position of Meliorchis within the extant orchid subtribe Goodyerinae. Using the ages of other fossil monocots and Meliorchis, we calibrated a molecular phylogenetic tree of the entire family Orchidaceae and showed that the most recent common ancestor of extant orchids lived in the Late Cretaceous, ~76-84 Myr ago. This finding suggests that the dramatic radiation of the orchid family took place during the Tertiary, shortly after the mass extinctions at the K/T boundary.

           The primary focus of my research is the evolutionary biology and ecology of bees. In particular, I am interested in understanding the co-adaptations and patterns of co-speciation that have resulted from long-term evolutionary associations between bee pollinators and their host plants. To address these questions, I am currently investigating host specialization and co-diversification among Neotropical euglossine bees and their orchid hosts.

Evolution of referential communication in eusocial bees.
Additionally, I am exploring various aspects of the ecology and evolution of referential communication in highly eusocial bees. Among bees, the highly eusocial corbiculate  bees (e.g. Apis and Melipona) are unique for their remarkable ability to use referential encoding to spatially inform nest mates the location of food sources. In collaboration with James C. Nieh and others, I am currently implementing comparative phylogenetic methods to investigate the origin, evolution and antiquity of referential communication in eusocial corbiculate Apidae.