KEW BULLETIN 62: 355–373 (2007) 355 A century of progress in grass systematics Khidir W. Hilu1 Summary. This paper presents an overview of progress in grass systematics with a focus on the past century and assesses its current status and future outlook. In concert with systematic biology, progress in grass systematics has gone through some leaps caused by the introduction of new approaches or emphasis on existing ones. Chromosome cytology, anatomy and chemistry provided useful information, but major recent contributions have come from advances in bioinformatics and molecular biology. Consequently, grass systematics has moved from an initial intuitive classification and phylogenetics to one incorporating analytical phenetic approaches, and culminating in the current stage of analytic phylogeny. As a result, a refined picture of grass phylogeny is emerging with good resolution at the base, but the tree lacks robustness in some places such as the monophyly of the “BEP” subfamilies and the relationships within the PACCAD clade. Systematic structure of a number of subfamilies is better understood now, but further studies are needed. With the rapid advancement in molecular systematic and bioinformatic tools, and in conjunction with a wealth of literature available on structural characters, a more refined picture of grass taxonomy and evolution is expected. However, caution needs to be exercised in our interpretations to avoid hasty decisions that can translate into regress rather than progress. This is an exciting time in the history of grass systematics and, undoubtedly, is a period of collaborative rather than individual effort. Key words. Poaceae, grasses, systematics, phylogenetics, history, evolution. Introduction valuable resources used in furthering our Although the c. 10,000 species in the Poaceae places it understanding of Poaceae systematics and in assessing as the fifth largest flowering plant family, a number of the mode and tempo of grass evolution. biological features raise it to a unique position. The Development of our views of grass systematics family has contributed crop species that provide followed major trends in advancement in the sciences about 80% of the annual global food (FAOSTAT and engineering. Major discoveries come in spurts 1999). In fact, four of the top crops that feed the and are usually followed by periods of relative world are cereal crops: wheat, rice, corn, and barley. stagnation. Optical development associated with This economic significance is matched by an microscopes provided a closer look at various ecological dominance as grasses cover about one fifth anatomical features; advancement in chemical of the earth (Shantz 1954). Grasslands are signature techniques revealed information on a number of of the Poaceae. Because of this economic and molecules such as flavonoids, proteins and alkaloids; ecological importance, grasses have attracted development of computers and various analytical considerable attention in the biological and software programs enhanced the capability of agricultural sciences. The past century showed handling large data sets and facilitated large-scale enormous amounts of research conducted on various analyses of available grass data; and finally, the biological aspects of the family, including systematics, current advancement in molecular biology and genetics, cytogenetics, breeding, physiology, anatomy, biotechnology provided us with the tools to look at developmental biology, chemistry, chromosome information derived from genes and genomes. structure, and, more recently, whole plastid and Parallel to, or as a consequence of, these major nuclear genome sequences. Results from these developments, the field of grass systematics underwent studies have provided a wealth of information on the a series of refinements that individually or collectively biology of the family, which have become and remain corresponded to the progressive stages cited above. Accepted for publication November 2006. 1 Department of Biological Sciences, Virginia Tech, Blacksburg, VA 24061, U.S.A. Phone: 1-540 231 6407; Fax: 1-540 552 9307; e-mail: [email protected] © The Board of Trustees of the Royal Botanic Gardens, Kew, 2007 356 KEW BULLETIN VOL. 62(3) These stages of progress in grass systematics, although Adanson’s (1763) division of the grass family into not necessary sharply distinct, are in general quite several “sections.” Outstanding contributions during recognizable and can be divided into two major that period included the allocation by Robert Brown periods, which in turn can be further divided into sub- (1814) of grass genera into the tribes “Panicaceae” and periods. The two main periods are the Predictive “Poaceae” (probably equivalent to what was later Period and the Analytical Period. These two periods recognized as subfamilies Panicoideae and the differ in the means by which hypotheses on grass traditional Festucoideae) based on detailed systematics are put forth. In the Predictive Period, understanding of spikelet structure. This system was taxonomic hypotheses were based on personal followed to a certain degree by Bentham (1881), and perceptions using available knowledge in the field and became the basis of the classification of Hitchcock personal conviction of the systematist. In the Analytical and Chase (1950) of the North American grasses into Period, hypotheses are based on analyses of available the subfamilies Festucoideae and Panicoideae. Between data with preset mathematical formulae that have the 1673 and the 1950 publications, a number of eventually taken advantage of computing systems. This contributions were made in which the number of paper will address grass systematics in these two subfamilies ranged between two and three periods, summarize the overall progress to date, assess (summarized in Hilu & Wright 1982 and GPWG the areas that need further work, and highlight 2001). The recognition of the major subgroups was possible future trends and needs in grass systematics. based largely on type of inflorescence, spikelet number and floret disarticulation. The emergence of additional characters in grass Predictive Period in Grass Systematics systematics had an early start with the recognition in This period began with the first publication on the the nineteenth century of useful patterns in starch classification of grasses by Adanson (1763). The grain (see Hilu & Wright 1982 and GPWG 2001). This period included both the classification of grasses into information attracted the attention to the presence of subfamilies and tribes (Predictive Taxonomy) as well important traits besides morphology and encouraged as the assessment of evolutionary patterns for the further work on these same traits or others by a new taxonomic units (Predictive Phylogeny). Although generation of researchers, ultimately leading to the two approaches overlap completely in time, it significant advances in grass systematics. These new would be best to address them separately. lines of research included (but were not necessarily limited to) microscopic features, chromosome Predictive Taxonomy cytology, leaf and stem anatomy, embryology, seedling Hilu & Wright (1982) recognized two stages in this type, photosynthetic pathways, seed storage protein, period, an initial stage that relied solely on flavonoid chemistry (see literature review in Hilu & morphological characters, and a second stage that Wright 1982; Clayton & Renvoize 1986; GPWG 2001). gradually incorporated microscopic structures and The Predictive Period began with what seems to be biochemical features. The initial stage started with simplistic view of grass classification and moved into Table 1. Selected taxonomic treatments of the Poaceae at higher taxonomic levels to represent the overall history and the various periods discussed here. Brown (1814) classification of grasses into two tribes roughly corresponds to later treatments at the subfamily. Brown 1814 Avdulov 1931 Stebbins & Caro 1982 Clayton & GPWG 2001 Crampton (1961) Renvoize 1986 Paniceae Poatae Bambusoideae Bambusoideae Bambusoideae Anomochlooideae Poaceae Sacchariferae Oryzoideae Streptochaetoideae Oryzoideae Pharoideae Arundinoideae Anomochlooideae Pooideae Puelioideae Festucoideae Olyroideae Chloridoideae Bambusoideae Eragrostoideae Centhostecoideae Panicoideae Ehrhartoideae Oryzoideae Arundinoideae Pooideae Ehrhartoideae Centothecoideae Aristidoideae Phragmitoideae Chloridoideae Festucoideae Centothecoideae Eragrostoideae Panicoideae Aristidoideae Danthonioideae Panicoideae Arundinoideae Micrairoideae © The Board of Trustees of the Royal Botanic Gardens, Kew, 2007 A CENTURY OF PROGRESS IN GRASS SYSTEMATICS 357 Fig. 1. Trends in Poaceae subfamily recognition since Brown (1814) treatment of the family. the impressive effort of exploring potential Roshevits (1937) provided what he termed “a informative characters in the Poaceae. These system of origin of grasses” based on the two- contributions clearly underscored the artificiality of subfamily system (Poatae and Sacchariferae) of Avdulov the two or three subfamily systems and pointed out (1930). Within the two subfamilies, three “series” the need for substantial taxonomic revisions. As a were recognized for the Poatae (Series Bambusiformes, consequence, new subfamilies were recognized (Table Phragmitiformes, and Festuciformes) and two for the 1) with the number varying from three (Avdulov Sacchariferae (Paniciformes and Eragrostiformes). He 1931) to 13 (Caro 1982). The highlight of the period depicted the Poatae emerging from