28
Michelle Waycott, Gabriele Procaccini, Donald H. Les and Thorsten B. H. Reusch
demonstrated that Cymodoceaceae, Posidoniaceae,
Ruppiaceae, Zannichelliaceae and Zosteraceae were
quite similar as a group, but differed considerably
from Hydrocharitaceae.
The comprehensive and highly influential work by
Cronquist (1981) allegedly was based on an integration of all data available at that time. That work depicted phylogenetic relationships within subclasses
and among orders using tree-like diagrams; however,
these were not produced using an explicit method.
Similar approaches were taken by authors of other
synthetic works (e.g. Dahlgren, 1980; Takhtajan,
1980). Comparable methods were used to develop
quite detailed classifications of seagrasses (e.g.
Tomlinson, 1982).
The incorporation of cladistic methodology led to
an important refinement in phylogenetic classification, that is it established an objective methodology
for reconstructing phylogenetic hypotheses. Unlike
previous approaches, cladistic analyses could provide explicit details regarding the character states
that defined particular relationships, which could
then be depicted as a hierarchical, tree-like diagram.
The first detailed cladistic analysis of seagrass
relationships was conducted by Dahlgren et al.
(Dahlgren and Rasmussen, 1983; Dahlgren, 1985)
as a part of their study of monocotyledons. Within
the subclass Alismatidae (which contains all seagrasses), they compared the distribution of states
for 56 morphological characters and provided the
first explicit evolutionary framework on which to
base a phylogenetic seagrass classification (Fig. 1).
Their results depicted seagrasses within three distinct clades: (1) Cymodoceaceae/Zannichelliaceae,
(2) Posidoniaceae/Zosteraceae and (3) Hydrocharitaceae (Ruppiaceae were merged with Potamogetonaceae). Within Hydrocharitaceae, Enhalus was
allied with Vallisneria (subfamily Vallisnerioideae),
whereas Halophila and Thalassia were each placed
in separate subfamilies (Halophiloideae, Thalassioideae). Dahlgren (1985) also argued against
the inclusion of Cymodoceaceae (incl. Amphibolis, Cymodocea, Halodule, Syringodium, Thalassodendron) within Zannichelliaceae (Althenia, Lepilaena, Zannichellia). Cox and Humphries (1993)
conducted a morphological cladistic analysis of
some seagrasses, which indicated the common origin of Posidoniaceae, Cymodoceaceae, and Zosteraceae. However, their analysis excluded many
non-seagrass families necessary to evaluate this
question adequately, thereby yielding unfounded results (see Les et al., 1997). Les and Haynes (1995)
reanalysed the Alismatidae data of Dahlgren (1985)
using a computer program capable of recovering additional maximum parsimony reconstructions and
also performed new analyses using corrected data.
Although the results were similar to Dahlgren’s original study, a number of discrepancies indicated that
morphological data alone could not be relied on to
provide compelling support for relationships in this
group.
Incorporation of molecular data in cladistic analyses provided the next major refinement in phylogenetic reconstructions of seagrass relationships. Les
et al. (1993) conducted preliminary cladistic analyses of Alismatidae using rbcL gene sequence data
from 8 families, but included only a few seagrass
genera, which did not provide much insight into their
relationships. In an rbcL analysis expanded to include 25 genera from all 15 alismatid families, Les
and Haynes (1995) began to disclose consistent
seagrass clades representing (1) Hydrocharitaceae,
(2) Posidoniaceae/Ruppiaceae/Cymodoceaceae, (3)
Zosteraceae, and (4) Zannichelliaceae (Lepilaena).
Seagrass relationships were evaluated further in
more comprehensive rbcL analyses of 55–69 alismatid species (eventually including 23 seagrass
species from all 14 marine genera) which specifically addressed the question of the number of seagrass origins (Waycott and Les, 1996; Les et al.,
1997) (Fig. 1).
The rbcL survey by Les et al. (1997) indicated
that marine angiosperms have evolved in at least
three separate lineages. Another independent lineage (Zannichelliaceae), which contains Lepilaena
marina, represents a fourth marine angiosperm origin given that this species (with several others in
the genus) is known to occupy marine habitats
(Womersley, 1984). To date, the study by Les et al.
(1997) remains the most complete molecular phylogenetic analysis of seagrass interrelationships at
the genus and family level. Procaccini et al. (1999b)
used trnL intron sequence data to study relationships among six marine genera in five families
and recovered distinct clades containing (1) Cymodocea/Posidonia/Ruppia, (2) Zostera/Phyllospadix
and (3) Halophila, thus in agreement with the rbcL
analysis by Les et al. (1997).
Molecular phylogenetic analyses have begun to
clarify some finer details of seagrass relationships.
First, the emerging pattern of phylogenetic relationships indicates that seagrasses should be divided
Michelle Waycott, Gabriele Procaccini, Donald H. Les and Thorsten B. H. Reusch
demonstrated that Cymodoceaceae, Posidoniaceae,
Ruppiaceae, Zannichelliaceae and Zosteraceae were
quite similar as a group, but differed considerably
from Hydrocharitaceae.
The comprehensive and highly influential work by
Cronquist (1981) allegedly was based on an integration of all data available at that time. That work depicted phylogenetic relationships within subclasses
and among orders using tree-like diagrams; however,
these were not produced using an explicit method.
Similar approaches were taken by authors of other
synthetic works (e.g. Dahlgren, 1980; Takhtajan,
1980). Comparable methods were used to develop
quite detailed classifications of seagrasses (e.g.
Tomlinson, 1982).
The incorporation of cladistic methodology led to
an important refinement in phylogenetic classification, that is it established an objective methodology
for reconstructing phylogenetic hypotheses. Unlike
previous approaches, cladistic analyses could provide explicit details regarding the character states
that defined particular relationships, which could
then be depicted as a hierarchical, tree-like diagram.
The first detailed cladistic analysis of seagrass
relationships was conducted by Dahlgren et al.
(Dahlgren and Rasmussen, 1983; Dahlgren, 1985)
as a part of their study of monocotyledons. Within
the subclass Alismatidae (which contains all seagrasses), they compared the distribution of states
for 56 morphological characters and provided the
first explicit evolutionary framework on which to
base a phylogenetic seagrass classification (Fig. 1).
Their results depicted seagrasses within three distinct clades: (1) Cymodoceaceae/Zannichelliaceae,
(2) Posidoniaceae/Zosteraceae and (3) Hydrocharitaceae (Ruppiaceae were merged with Potamogetonaceae). Within Hydrocharitaceae, Enhalus was
allied with Vallisneria (subfamily Vallisnerioideae),
whereas Halophila and Thalassia were each placed
in separate subfamilies (Halophiloideae, Thalassioideae). Dahlgren (1985) also argued against
the inclusion of Cymodoceaceae (incl. Amphibolis, Cymodocea, Halodule, Syringodium, Thalassodendron) within Zannichelliaceae (Althenia, Lepilaena, Zannichellia). Cox and Humphries (1993)
conducted a morphological cladistic analysis of
some seagrasses, which indicated the common origin of Posidoniaceae, Cymodoceaceae, and Zosteraceae. However, their analysis excluded many
non-seagrass families necessary to evaluate this
question adequately, thereby yielding unfounded results (see Les et al., 1997). Les and Haynes (1995)
reanalysed the Alismatidae data of Dahlgren (1985)
using a computer program capable of recovering additional maximum parsimony reconstructions and
also performed new analyses using corrected data.
Although the results were similar to Dahlgren’s original study, a number of discrepancies indicated that
morphological data alone could not be relied on to
provide compelling support for relationships in this
group.
Incorporation of molecular data in cladistic analyses provided the next major refinement in phylogenetic reconstructions of seagrass relationships. Les
et al. (1993) conducted preliminary cladistic analyses of Alismatidae using rbcL gene sequence data
from 8 families, but included only a few seagrass
genera, which did not provide much insight into their
relationships. In an rbcL analysis expanded to include 25 genera from all 15 alismatid families, Les
and Haynes (1995) began to disclose consistent
seagrass clades representing (1) Hydrocharitaceae,
(2) Posidoniaceae/Ruppiaceae/Cymodoceaceae, (3)
Zosteraceae, and (4) Zannichelliaceae (Lepilaena).
Seagrass relationships were evaluated further in
more comprehensive rbcL analyses of 55–69 alismatid species (eventually including 23 seagrass
species from all 14 marine genera) which specifically addressed the question of the number of seagrass origins (Waycott and Les, 1996; Les et al.,
1997) (Fig. 1).
The rbcL survey by Les et al. (1997) indicated
that marine angiosperms have evolved in at least
three separate lineages. Another independent lineage (Zannichelliaceae), which contains Lepilaena
marina, represents a fourth marine angiosperm origin given that this species (with several others in
the genus) is known to occupy marine habitats
(Womersley, 1984). To date, the study by Les et al.
(1997) remains the most complete molecular phylogenetic analysis of seagrass interrelationships at
the genus and family level. Procaccini et al. (1999b)
used trnL intron sequence data to study relationships among six marine genera in five families
and recovered distinct clades containing (1) Cymodocea/Posidonia/Ruppia, (2) Zostera/Phyllospadix
and (3) Halophila, thus in agreement with the rbcL
analysis by Les et al. (1997).
Molecular phylogenetic analyses have begun to
clarify some finer details of seagrass relationships.
First, the emerging pattern of phylogenetic relationships indicates that seagrasses should be divided