As a systematic botanist, I have been working to address questions regarding the evolutionary relationships of taxa and the evolution of specific traits associated with those taxa.  My dissertation research has focused on members of the mustard plant family (Brassicaceae), which encompasses over 340 genera and 3500 spp.  The family contains important crop species, such as broccoli, brussels sprouts, cauliflower, kale, horseradish, mustards, as well as horticultural species and the molecular biology model organism Arabidopsis thaliana.  Despite the economic and scientific importance of the family, intrafamilial relationships remain relatively uncertain.  This is due to taxonomic reliance on a few morphological characters that appear to display high degrees of convergence and parallelism in combination with a lack of discrete morphological variation available to resolve intrafamilial relationships.
        I have been working to help resolve the family’s systematic difficulties through phylogenetic studies at higher and lower taxonomic levels, by providing specimens from poorly collected geographic regions, through the incorporation of a previously unused DNA sequence data source, and the simultaneous analysis of morphological and molecular data.  I have focused this project on the poorly know Mexican genus Sphaerocardamum (8 species) and its close relatives in the “Halimolobos clade” (Halimolobos, Mancoa, Pennellia, and Sphaerocardamum).  The specific questions I am addressing in this group include:  1) the position of Sphaerocardamum and its relatives within the family, 2) the naturalness of Sphaerocardamum, 3) an initial test of naturalness for Halimolobos, Mancoa, and Pennellia, 4) morphological character evolution within the Halimolobos clade, and 5) genome size evolution in Sphaerocardamum.

        Below is an overview of the approaches taken to answer these questions and current results:

1)  Collaborative Brassicaceae family level analyses:  Family level phylogenetic studies (evolutionary) have been conducted in collaboration with Dr. Robert Price (Athens GA, USA) using new additions to his existing family level DNA sequence dataset (chloroplast ndhF).  We have used this collaborative study to initially test the naturalness (monophyly) of Sphaerocardamum and to identify close relatives of the genus.(Price and Bailey, in prep).  Our results identified it as a member of a New World mostly Mexican species group that includes more than 40 species and 4 genera (Halimolobos, Mancoa, Pennellia, and Sphaerocardamum).  These species are more closely related to Arabidopsis and members of the genus Arabis than they are to true Brassicoid mustards (Brassica).  In addition, they display a striking variety of fruit morphologies in a lineage of closely related species.

2)  Lower level phylogenetic studies:  After Dr. Price and I established the general position of the genus within the family, I began a more thorough study of Sphaerocardamum and its relatives.  This study was conducted to fully test naturalness of Sphaerocardamum and as an initial test of naturalness for the other three genera in the clade.  In addition, the study was designed to comment on the biogeography and evolution of fruit types in this diverse New World lineage.  This portion of the project has been accomplished by analysis of DNA sequence data from three potentially independent DNA sequence datasets (nrDNA ITS, pistillata intron 1, and trnL intron) and morphology (Bailey et al., in prep).  The results support the naturalness of Sphaerocardamum, however, Halimolobos and Mancoa clearly represent unnatural assemblages that will require further study in order to more fully understand their generic boundaries (see project proposal).

3)  Monographic treatment:  Beyond studying the phylogenetic relationships among these species, I  have focused on species delimitation issues in Sphaerocardamum using data from anatomy, crossing studies, cytology (including flow cytometry), DNA sequences, and morphology.  The addition of numerous new collections has helped to identify continuous variation between previously recognized species.  I currently recognize four valid species in the genus (reduced from the eight previously recognized).  All of the information generated through the study of Sphaerocardamum will be summarized in a monographic treatment addressing the classification, biology, and evolution of the genus (Bailey and Doyle, in prep.).

4)  Incorporating new sources of data: At the beginning of this project, I developed primers to test the phylogenetic utility of a previously unused DNA sequence data source in phylogenetics.  The MADS-box gene pistillata was sampled from Sphaerocardamum and other Brassicaceae to test the utility of its ca. 1000 bp first intron.  Linkage information in Brassica and Arabidopsis suggest that this sequence is unlinked to the other two DNA loci being used (nrDNA ITS and cptrnL intron).  The results of this study indicate that this single to low-copy nuclear gene can be appropriate for use in Brassicaceae phylogenetics (Bailey and Doyle, 1999).

Bailey, C.D. and J.J. Doyle.  1999.  Potential phylogenetic utility of the low-copy nuclear gene pistillata in dicotyledonous plants:  comparison to nrDNA ITS and trnL intron in Sphaerocardamum and other Brassicaceae.  Molecular Phylogenetics and Evolution 13: 20-30.