6
C. den Hartog and J. Kuo
In our opinion the three Australian species are in
general well distinguished by morphological differences in their leaf-tip and nervation. The shape of the
leaf apex, in spite of having considerable variation,
still is one of most important vegetative characters
for identification of these Zostera species. Further
Z. capricorni has leaves with five longitudinal
nerves, while the other two species have only three.
An other important fact is, that the three species
have each a different area of distribution with only
marginal overlap. In these areas of overlap intermediate forms occur, but these may be the result
of hybridization. We certainly do not deny that the
species are closely related. Robertson (1984) noticed
also that a broad spectrum of intergrades occurs and
recommended further basic taxonomic work to elucidate the Z. mucronata–Z. muelleri–Z . capricorni
complex. Turner et al. (1996) also could not decide
the species status of their studied Zostera material
from New Zealand.
However, Les et al. (2002) echoed Robertson’s
(1984) suggestions recommending ‘common garden
experiments to be conducted among these species’
to see whether those leaf tip and other morphological characters are the result of environmental conditions, or are genetically determined, or both. Until such fundamental data are available, we recommend here that at least Z. muelleri, Z. capricorni,
and Z. mucronata should continue to be recognized.
McMillan (1982) investigated five Zostera species,
amongst which four members of the subgenus
Zosterella, and found that each species presented a
different isozyme pattern.
Further, it should be noted that three out of 31
characters separate Z. noltii and Z. japonica (see Les
et al., 2002), but the matK tree shows a close affinity
of these two species, which have disjunct distributions (Tanaka et al., 2003).
Robertson (1984) recognized two ecotypes of
Zostera muelleri from south-eastern Australia, one
in the intertidal belt of sheltered bays, and a more
robust estuarine form in lagoons and more or less
land-locked waters, which is almost permanently
submerged. She stressed that numerous intergrades
occur between them. Similar observations have been
made for other Zostera species, e.g. Z. noltii in western Europe, Mauritania, and the Mediterranean, Z.
capricorni in New South Wales, and Z. capensis in
South Africa.
Phyllospadix Hooker, Fl. Bor. Am. 2 (1838) 171
Type species: Phyllospadix scouleri Hooker
The genus contains five species, all occurring along
the northern temperate coasts of the Pacific. The
genus is rather homogeneous, although two groups
of species can be recognized. One group consists of
the two species, P. scouleri and P. torreyi, and occurs along the west coasts of Canada, the USA, and
Baia California. It is characterized by having rhizome nodes with 6–10 roots (in two rows), leaves
with only three nerves (P. japonicus belonging to
the other group also has three nerves), and generative axes consisting of one to several internodes
and bearing one or more pedunculate spathes. The
second group has three species, distributed in cold
temperate waters in eastern Asia (P. iwatensis and
P. japonicus) and North America from the Aleutic
Islands southward to Oregon (P. serrulatus). In these
three species the rhizome nodes have only two roots,
and the generative axes are reduced to short pedunculate spathes.
Tsvelev (1981) erected a special section, Phyllospadix sect. Sagitticarpus, to include P. torreyi, because its inflorescence produces numerous spathes,
without considering the infrageneric classification
of the other species of the genus. Although we can
recognize within the genus two species groups there
is in our opinion no reason to subdivide Phyllospadix
into sections or subgenera. If these groups have to
be formally upgraded to the section level, the group
which contains the type species should be named in
accordance with the rules of the botanical nomenclature P. sect. Phyllospadix and P. sect. Sagitticarpus
is in that case a superfluous synonym.
Further, it should be mentioned here that Tsvelev
(1981) described two new species, P. juzepczukii
and P. ruprechtii from eastern Russia and California
respectively. Unfortunately, we have not been able
to see material from these species. It appears that
the leaves of P. juzepczukii have only three nerves,
as in P. japonicus, and these taxa may possibly be
identical, but in that case the species would show
a very remarkable disjunct area of distribution. As
far is known to us P. japonicus is restricted to the
Honshu coast of the Japanese Sea, but is absent
from Hokkaido, Korea and China. P. juzepczukii
seems to be widely distributed in the northern Far
East. It is clear from the diagnosis of P. ruprechtii
that this species is synonymous with P. scouleri.
Finally, Dawson et al. (1960) have found some 3nerved, almost perfectly cylindrical leaf fragments
C. den Hartog and J. Kuo
In our opinion the three Australian species are in
general well distinguished by morphological differences in their leaf-tip and nervation. The shape of the
leaf apex, in spite of having considerable variation,
still is one of most important vegetative characters
for identification of these Zostera species. Further
Z. capricorni has leaves with five longitudinal
nerves, while the other two species have only three.
An other important fact is, that the three species
have each a different area of distribution with only
marginal overlap. In these areas of overlap intermediate forms occur, but these may be the result
of hybridization. We certainly do not deny that the
species are closely related. Robertson (1984) noticed
also that a broad spectrum of intergrades occurs and
recommended further basic taxonomic work to elucidate the Z. mucronata–Z. muelleri–Z . capricorni
complex. Turner et al. (1996) also could not decide
the species status of their studied Zostera material
from New Zealand.
However, Les et al. (2002) echoed Robertson’s
(1984) suggestions recommending ‘common garden
experiments to be conducted among these species’
to see whether those leaf tip and other morphological characters are the result of environmental conditions, or are genetically determined, or both. Until such fundamental data are available, we recommend here that at least Z. muelleri, Z. capricorni,
and Z. mucronata should continue to be recognized.
McMillan (1982) investigated five Zostera species,
amongst which four members of the subgenus
Zosterella, and found that each species presented a
different isozyme pattern.
Further, it should be noted that three out of 31
characters separate Z. noltii and Z. japonica (see Les
et al., 2002), but the matK tree shows a close affinity
of these two species, which have disjunct distributions (Tanaka et al., 2003).
Robertson (1984) recognized two ecotypes of
Zostera muelleri from south-eastern Australia, one
in the intertidal belt of sheltered bays, and a more
robust estuarine form in lagoons and more or less
land-locked waters, which is almost permanently
submerged. She stressed that numerous intergrades
occur between them. Similar observations have been
made for other Zostera species, e.g. Z. noltii in western Europe, Mauritania, and the Mediterranean, Z.
capricorni in New South Wales, and Z. capensis in
South Africa.
Phyllospadix Hooker, Fl. Bor. Am. 2 (1838) 171
Type species: Phyllospadix scouleri Hooker
The genus contains five species, all occurring along
the northern temperate coasts of the Pacific. The
genus is rather homogeneous, although two groups
of species can be recognized. One group consists of
the two species, P. scouleri and P. torreyi, and occurs along the west coasts of Canada, the USA, and
Baia California. It is characterized by having rhizome nodes with 6–10 roots (in two rows), leaves
with only three nerves (P. japonicus belonging to
the other group also has three nerves), and generative axes consisting of one to several internodes
and bearing one or more pedunculate spathes. The
second group has three species, distributed in cold
temperate waters in eastern Asia (P. iwatensis and
P. japonicus) and North America from the Aleutic
Islands southward to Oregon (P. serrulatus). In these
three species the rhizome nodes have only two roots,
and the generative axes are reduced to short pedunculate spathes.
Tsvelev (1981) erected a special section, Phyllospadix sect. Sagitticarpus, to include P. torreyi, because its inflorescence produces numerous spathes,
without considering the infrageneric classification
of the other species of the genus. Although we can
recognize within the genus two species groups there
is in our opinion no reason to subdivide Phyllospadix
into sections or subgenera. If these groups have to
be formally upgraded to the section level, the group
which contains the type species should be named in
accordance with the rules of the botanical nomenclature P. sect. Phyllospadix and P. sect. Sagitticarpus
is in that case a superfluous synonym.
Further, it should be mentioned here that Tsvelev
(1981) described two new species, P. juzepczukii
and P. ruprechtii from eastern Russia and California
respectively. Unfortunately, we have not been able
to see material from these species. It appears that
the leaves of P. juzepczukii have only three nerves,
as in P. japonicus, and these taxa may possibly be
identical, but in that case the species would show
a very remarkable disjunct area of distribution. As
far is known to us P. japonicus is restricted to the
Honshu coast of the Japanese Sea, but is absent
from Hokkaido, Korea and China. P. juzepczukii
seems to be widely distributed in the northern Far
East. It is clear from the diagnosis of P. ruprechtii
that this species is synonymous with P. scouleri.
Finally, Dawson et al. (1960) have found some 3nerved, almost perfectly cylindrical leaf fragments