Chapter 2 Seagrass Genetics and Evolution
27
Table 1. Comparison of selected marine angiosperm (seagrass) classifications.
Marine
Ascherson and Arber Sculthorpe Den Hartog Cronquist Tomlinson Dahlgren
Thorne Les et al.
genera
Graebner (1907) (1920) (1967)
(1970)
(1981)
(1982)
et al. (1985) (1992) (1997)
Enhalus
HYD
HYD HYD
HYD
HYD
HYD
HYD
HYD HYD
Halophila
HYD
HYD HYD
HYD
HYD
HYD
HYD
HYD HYD
Thalassia
HYD
HYD HYD
HYD
HYD
HYD
HYD
HYD HYD
Amphibolis
POT
POT ZAN
POT
CYM
CYM
CYM
CYM CYM
Cymodocea
POT
POT ZAN
POT
CYM
CYM
CYM
CYM CYM
Halodule
POT
POT ZAN
POT
CYM
CYM
CYM
CYM CYM
Heterozostera
–
–
ZOS
POT
ZOS
ZOS
ZOS
ZOS
ZOS
Lepilaena
a
POT
–
ZAN
–
ZAN
ZAN
ZAN
ZAN
ZAN
Phyllospadix
POT
POT ZOS
POT
ZOS
ZOS
ZOS
ZOS
ZOS
Posidonia
POT
POT POS
POT
POS
POS
POS
POS
POS
Ruppia
POT
POT RUP
POT
RUP
POT
POT
POT
RUP
Syringodium
POT
–
ZAN
POT
CYM
CYM
CYM
CYM CYM
Thalassodendron POT
–
–
POT
CYM
CYM
CYM
CYM CYM
Zostera
POT
POT ZOS
POT
ZOS
ZOS
ZOS
ZOS
ZOS
a Not identified as marine, but included in Zannichelliaceae by Les et al. (1997).
CYM = Cymodoceaceae; HYD = Hydrocharitaceae; POS = Posidoniaceae; POT = Potamogetonaceae; RUP = Ruppiaceae;
ZAN = Zannichelliaceae; ZOS = Zosteraceae. — = not considered by author.
Thalassia and Halophila within Hydrocharitaceae
(Richard, 1811; Rohrbach, 1871; Balfour, 1879).
Bentham and Hooker (1883) and Ascherson and
Graebner (1907) maintained the division of seagrass
species into separate families (Hydrocharitaceae,
Potamogetonaceae). This arrangement was later
adopted by Arber (1920), and the distinction of
seagrass species in the Hydrocharitaceae has been
followed since. Yet, even though seagrasses were
divided among separate families, they were believed
by some to retain a relatively close relationship.
Balfour (1879) considered Halophila as the link
between the families Potamogetonaceae and Hydrocharitaceae, Arber (1920) and Setchell (1920)
regarded these same families as ‘nearly related’.
With the exception of den Hartog (1970), who
simply adopted Ascherson and Graebner’s (1907)
two family system, classifications increasingly partitioned seagrass genera among several additional
families by the later part of the 20th Century
(Table 1). den Hartog (1970) did, however, remark
that Ascherson and Graebner’s concept of Potamogetonaceae was heterogeneous and needed to be subdivided. Major similarities and differences in the
historical family groupings of seagrass genera are
shown by the selected classifications in Table 1. It
is noteworthy that greater taxonomic consistency
underlies these examples, because Ascherson and
Graebner’s dual family system (followed by Arber
and den Hartog) subdivided Potamogetonaceae
into tribes corresponding to Cymodoceaceae (Cymodoceeae), Posidoniaceae (Posidonieae), Potamogetonaceae (Potamogetoneae), Zannichelliaceae
(Zannichellieae) and Zosteraceae (Zostereae), thus
yielding the same six taxa recognized by many of the
later classifications. These examples illustrate that a
relatively stable seagrass classification has existed
for more than a century.
However, the nature of scientific evidence in support of the assorted marine angiosperm classifications shown in Table 1 is quite variable. The earliest classifications essentially represented opinions
of natural groups based on an author’s perception
of which taxa were most similar morphologically
and anatomically. The incorporation of evolutionary
theory in the 20th Century inspired the adoption of
classifications that also strived to be meaningful phylogenetically with most modern classifications now
being implicitly phylogenetic (Judd et al., 2002).
In early phylogenetically based studies, relationships were ascertained by identifying groups that
shared a number of distinctive features, arguably
indicating their common ancestry. Bessey (1915)
incorporated the concept of character state polarity to help differentiate groups as relatively primitive or advanced. Otherwise, this approach did not
differ materially from that used to establish putatively ‘natural’ classifications other than assuming that similarity resulted from a common evolutionary ancestry. Efforts to better quantify group
similarity led to the incorporation of ‘phenetic’ approaches, which performed cluster analyses of character matrices using computerized algorithms. Such
an approach was taken by Clifford (1970) who
27
Table 1. Comparison of selected marine angiosperm (seagrass) classifications.
Marine
Ascherson and Arber Sculthorpe Den Hartog Cronquist Tomlinson Dahlgren
Thorne Les et al.
genera
Graebner (1907) (1920) (1967)
(1970)
(1981)
(1982)
et al. (1985) (1992) (1997)
Enhalus
HYD
HYD HYD
HYD
HYD
HYD
HYD
HYD HYD
Halophila
HYD
HYD HYD
HYD
HYD
HYD
HYD
HYD HYD
Thalassia
HYD
HYD HYD
HYD
HYD
HYD
HYD
HYD HYD
Amphibolis
POT
POT ZAN
POT
CYM
CYM
CYM
CYM CYM
Cymodocea
POT
POT ZAN
POT
CYM
CYM
CYM
CYM CYM
Halodule
POT
POT ZAN
POT
CYM
CYM
CYM
CYM CYM
Heterozostera
–
–
ZOS
POT
ZOS
ZOS
ZOS
ZOS
ZOS
Lepilaena
a
POT
–
ZAN
–
ZAN
ZAN
ZAN
ZAN
ZAN
Phyllospadix
POT
POT ZOS
POT
ZOS
ZOS
ZOS
ZOS
ZOS
Posidonia
POT
POT POS
POT
POS
POS
POS
POS
POS
Ruppia
POT
POT RUP
POT
RUP
POT
POT
POT
RUP
Syringodium
POT
–
ZAN
POT
CYM
CYM
CYM
CYM CYM
Thalassodendron POT
–
–
POT
CYM
CYM
CYM
CYM CYM
Zostera
POT
POT ZOS
POT
ZOS
ZOS
ZOS
ZOS
ZOS
a Not identified as marine, but included in Zannichelliaceae by Les et al. (1997).
CYM = Cymodoceaceae; HYD = Hydrocharitaceae; POS = Posidoniaceae; POT = Potamogetonaceae; RUP = Ruppiaceae;
ZAN = Zannichelliaceae; ZOS = Zosteraceae. — = not considered by author.
Thalassia and Halophila within Hydrocharitaceae
(Richard, 1811; Rohrbach, 1871; Balfour, 1879).
Bentham and Hooker (1883) and Ascherson and
Graebner (1907) maintained the division of seagrass
species into separate families (Hydrocharitaceae,
Potamogetonaceae). This arrangement was later
adopted by Arber (1920), and the distinction of
seagrass species in the Hydrocharitaceae has been
followed since. Yet, even though seagrasses were
divided among separate families, they were believed
by some to retain a relatively close relationship.
Balfour (1879) considered Halophila as the link
between the families Potamogetonaceae and Hydrocharitaceae, Arber (1920) and Setchell (1920)
regarded these same families as ‘nearly related’.
With the exception of den Hartog (1970), who
simply adopted Ascherson and Graebner’s (1907)
two family system, classifications increasingly partitioned seagrass genera among several additional
families by the later part of the 20th Century
(Table 1). den Hartog (1970) did, however, remark
that Ascherson and Graebner’s concept of Potamogetonaceae was heterogeneous and needed to be subdivided. Major similarities and differences in the
historical family groupings of seagrass genera are
shown by the selected classifications in Table 1. It
is noteworthy that greater taxonomic consistency
underlies these examples, because Ascherson and
Graebner’s dual family system (followed by Arber
and den Hartog) subdivided Potamogetonaceae
into tribes corresponding to Cymodoceaceae (Cymodoceeae), Posidoniaceae (Posidonieae), Potamogetonaceae (Potamogetoneae), Zannichelliaceae
(Zannichellieae) and Zosteraceae (Zostereae), thus
yielding the same six taxa recognized by many of the
later classifications. These examples illustrate that a
relatively stable seagrass classification has existed
for more than a century.
However, the nature of scientific evidence in support of the assorted marine angiosperm classifications shown in Table 1 is quite variable. The earliest classifications essentially represented opinions
of natural groups based on an author’s perception
of which taxa were most similar morphologically
and anatomically. The incorporation of evolutionary
theory in the 20th Century inspired the adoption of
classifications that also strived to be meaningful phylogenetically with most modern classifications now
being implicitly phylogenetic (Judd et al., 2002).
In early phylogenetically based studies, relationships were ascertained by identifying groups that
shared a number of distinctive features, arguably
indicating their common ancestry. Bessey (1915)
incorporated the concept of character state polarity to help differentiate groups as relatively primitive or advanced. Otherwise, this approach did not
differ materially from that used to establish putatively ‘natural’ classifications other than assuming that similarity resulted from a common evolutionary ancestry. Efforts to better quantify group
similarity led to the incorporation of ‘phenetic’ approaches, which performed cluster analyses of character matrices using computerized algorithms. Such
an approach was taken by Clifford (1970) who