Ruppia maritima Posldonla oceanica Posidonia kirkmanii
Length 137 CI 0.949 Rl 0.949 RC 0.901
Halodule uninervis Karumba N Austraiia Halodule uninervis Townsville e Austraiia Halodule pinifolia Malaysia Halodule pinifolia North Carolina, USA Halodule wrightii Florida Bay, USA ■ Halodule wrightii Bermuda Halodule wrightii Bermuda Halodule wrightii Florida Bay, USA J Halodule wrightii Bocus del Toro, Panama 13 Halodule wrightii San Blas, Panama J Halodule uninervis Townsviite E Austraiia Halodule pinifolia Whitsundays £ Austraiia
Fig. 2. (a) Above ITS and (b) below trnL parsimony phylogram of the genus Halodule (bootstrap support shown adjacent to nodes), localities of collections shown next to species name (Waycott and Barnes unpublished).
distinct (the latter two representing closely related sister species); however, the five remaining Australian taxa (P. coriacea, P. ostenfeldii, P. robertsonii, P. denhartogii, P. kirkmanii) appear to represent only minor variants of a single species and are indistinct both morphologically and genetically.
A phylogenetic study of Halophila has been conducted by Waycott et al. (2002) who analysed ITS data from 36 accessions of 11 recognized species.
ITS (nrDNA) and trnL (cpDNA) combined
Hal od ule wrightü
Halo du le uninervis
1759 bp Length 449 CI 0.924 Rl 0.867 RC 0.802
Posidonia oceanica P. angustifolia P. australis _ P. sinuosa P. coriacea P. kirkmanii P. robertsonii P. denhartogii P. ostenfeldii
Fig. 3. Combined data of ITS and trnL parsimony phylogram of Posidonia (bootstrap support shown adjacent to nodes), localities of collections shown next to species name (Waycott and Les unpublished).
Their results dispute the taxonomic limits of H. ovalis, wherein some populations are closely related to H. australis. The rare H. johnsonii and H. hawaiiana were not separable from H. ovalis by ITS data and perhaps represent only minor variants of a single species. However, most other species appear to be distinct and are associated with a pattern of vegetative reduction proceeding phylogenetically from complex leaf arrangements to reduced, simplified phyllotaxy (Fig. 4). Two Halophila species (H. capricorni, H. baillonis) remain unsurveyed for ITS. It has also been observed that rDNA can accumulate pseudogenes, making the verification of sequences important before use in analyses of relationships (Ruggiero and Procaccini, 2004). Additional loci (especially cpDNA) should be surveyed to seek further support for relationships disclosed by the initial ITS data analysis.
Systematic relationships within Phyllospadix (five species) have not been investigated in any detail. Isozyme patterns have been compared for three species, with higher similarity reported between P. scouleri and P. torreyi than between either species or P. serrulatus (Triest, 1991b). Phyllospadix would benefit from a thorough evaluation of taxo-nomic limits and relationships using both molecular and morphological data sets.
Relationships within Zostera (nine species including Heterozostera) have now been studied in some detail. Chromosome morphology is distinct between subgenera with those of subgenus Zostera smaller than those of subgenus Zosterella (Uchiyama, 1996). Similarly, isozymes show different patterns between Z. marina (subg. Zostera) and species of subgenus Zosterella; however, only slight isozymic differences were observed among Z. capensis, Z. capri-corni, Z. muelleri, and Z. novazelandica (subgenus Zosterella) (Triest, 1991b). Uchiyama (1996) conducted a molecular analysis of three Zostera species using 18s RDNA RFLP data. His results also showed differences between species of different subgenera, but sampling was insufficient to address relationships in detail. Les et al. (1997) surveyed three Zostera species (representing both subgenera) in their rbcL analysis of Alismatidae. Again, species from the different subgenera were considerably divergent. Les et al. (1997) also provided evidence that the taxon formerly recognized as a separate genus (Heterozostera tasmanica) falls within the genus Zostera phylogenetically and should be included within Zostera. (Les et al., 2002) performed a morphological phylogenetic analysis of all Zostera species (including Heterozostera) and also evaluated systematic relationships among eight taxa using a combined data set consisting of DNA sequences from ITS, rbcL, and the trnK intron. Results of this study reinforced earlier work that indicated significant morphological and molecular divergence between the two subgenera of Zostera (Fig. 5). However, there was no phylogenetically defensible structure to accessions sampled for Z. capensis, Z. capricorni, Z. muelleri, and Z. novazelandica, leading to the recognition of only one variable species (Z. muelleri) in that group. This study verified that Heterozostera should be merged with Zostera. Recently, Les and Moody (unpublished) obtained ITS sequences from an extremely broad-leaved accession of Zostera from California (USA), which has been referred to as Zostera latifolia by some authors. Setchell (1927) regarded Z. latifolia to be an ecological variant of Z. marina. Les and Moody detected only a single substitution in the entire ITS region between the broad-leaved form from California and narrow-leaved material of Zostera marina from the east coast (Connecticut, USA), evidence that supports Setchell's merger of these taxa, despite conspicuous morphological difference in leaf size. These results are supported by the work of Tanaka
et al. (2003) who analysed the chloroplast gene region matK. Their results suggest that Z. noltii and Z. japonica are sister to the combined taxon Z. muel-leri (=Z. capricorni of Les et al., 2002). Further exploration of the intrageneric relationships would be useful to provide insight into suggestions of creating additional genera within the family by Tomlinson and Posluszny (2001).
The taxonomy and systematics of Ruppia is in serious need of study. Until recently, Ruppia had been placed within the family Potamogetonaceae, usually as a separate subfamily. Les et al. (1993) showed that rbcL data placed Ruppia closer to Cymodoceaceae (Syringodium). The alliance of Cy-modoceaceae, Ruppiaceae, and Posidoniaceae was verified in later, more comprehensive studies (Waycott and Les, 1996; Les et al., 1997). The actual number of Ruppia species is not known with any certainty. Cook (1996) reported 2-10 species worldwide. The study by Les et al. (1997) included only
two species (R. maritima from North America, R. megacarpa from Australia) whose rbcL sequences were fairly distinct. Guha and Mondal (1999) studied pollen morphology in Ruppia and concluded a worldwide revision of Ruppia which includes a full appraisal of morphological characters as well as molecular phylogenetic analyses will be necessary before further systematic details such as species boundaries and relationships can be ascertained in this genus.
Lepilaena (five species; Zannichelliaceae) also has not been studied systematically in any great detail. Molecular (rbcL) data clearly showed Lepilaena australis as related to Zannichellia, but other species await study. Two species (Lepilaena marina; L. cylin-drocarpa) are truly marine (Womersley, 1984) and it would be informative to determine whether the marine habit is basal or derived in this genus of otherwise freshwater species. Although quite likely, the monophyly of Lepilaena has not yet been verified by phylogenetic analysis.
These studies have all verified the utility of detailed analysis of the intraspecific variation across a wide geographic range for seagrass species using DNA sequence data and careful phylogenetic analysis. Broader scale studies may require considerable effort in obtaining samples from extremes of the range of species to better describe the finer scale evolutionary trends within genera and species. These types of study begin to impinge upon the traditional population genetic approach to understanding relat-edness of seagrass populations as will be discussed in the following sections. The intersection of these two fields of study represents the investigation of species phylogeography (e.g. for review see Avise, 2000). This field of research has barely been touched in seagrasses and while at present markers that adequately detect the historical biogeographic processes are unavailable (Schaal et al., 1998), future research will proved invaluable to the study of broad scale evolution of seagrass species. The field of molecular systematics is undergoing continuous and rapid development and this will provide fertile ground for future research activities.
Was this article helpful?