Geoscience Reference
In-Depth Information
because the parsimony criterion no longer suggested
that they should have been restricted to freshwater
(Fig. 1c). The intolerance to saltwater, which
characterizes most lissamphibians (but not all; see
Schmidt 1957; Garland et al. 1997), could have
appeared as early as the first amphibian or as late
as the last common ancestor of all lissamphibians.
Schultze (1985) had already reached similar con-
clusions, and Milner (1987, pp. 500 - 501) stated
that
'lepospondyls', embolomeres and seymouria-
morphs). This phylogeny is probably less supported
than other recent alternatives because it conflicts
with all published molecular phylogenies (and
most morphological ones) which support the mono-
phyly of Lissamphibia with respect to Amniota
(Laurin 2002).
Other recent phylogenies do not fit the patterns
presented above. For instance, McGowan (2002)
suggests that many 'lepospondyls' are part of the
amphibian crown (and hence should have been
mostly freshwater forms) but, since his analysis
does not include amniotes or reptiliomorphs, it is
impossible to determine whether temnospondyls
are stem-amphibians or stem-tetrapods under his
proposal. His phylogeny suggests that most 'lepos-
pondyls' did not tolerate salt or brackish water,
but makes no prediction about environmental pre-
ferences of temnospondyls. This phylogeny was
recently shown to be based on questionable anatom-
ical interpretations (Marjanovi´ & Laurin 2008b,
2009).
To summarize, in the context of phylogenies pro-
posed early in the 20th century, parsimony
suggested (Fig. 1) that Palaeozoic amphibians did
not tolerate saltwater (Fig. 1a, b). More recent phy-
logenies usually suggest that Lissamphibia excludes
all or most Palaeozoic amphibians (Fig. 1c - e),
which implies no intolerance (but not necessarily
tolerance either) to saltwater in amphibians
(Fig. 1c - f). These trees suggest that some taxa tra-
ditionally attributed to Amphibia are stem-tetrapods
(Fig. 1d) or reptiliomorphs (Fig. 1e), and this
increases further the number of taxa which can be
expected to have tolerated saltwater.
...most recent workers believe the living amphibians
to form a clade (usually referred to as the Lissamphi-
bia) definable by a series of unique characteristics,
most of which are not known in any Palaeozoic
amphibian-grade tetrapod (see Rage & Janvier 1982
for a recent discussion). It can thus no longer be
assumed that the freshwater dependence of most
living amphibians is an inheritance from the Palaeo-
zoic amphibian-grade tetrapods; it may represent a
specialization acquired later in the early stages of
lissamphibian evolution.
Several recent and comprehensive computer-
assisted phylogenetic analyses of tetrapods suggest
that temnospondyls, formerly thought to be early
amphibians, are stem-tetrapods (Laurin & Reisz
1997, 1999; Laurin 1998a; Anderson 2001; Vallin
& Laurin 2004; Marjanovi´ & Laurin 2009). There-
fore, the parsimony criterion actually suggests that
this group tolerated saltwater (Fig. 1d); some tem-
nospondyls may have lived in freshwater and
others probably inhabited the coastal marine
environment at least during juvenile and adult
stages (see also Schult 1994; Schultze et al. 1994).
The discovery of well-preserved remains of the tem-
nospondyl Iberospondylus in a coastal environment
(Laurin & Soler-Gij ´n 2001, 2006) should not there-
fore be viewed as anomalous. However, this phylo-
geny makes no prediction about habitat preference
in early amphibians ('lepospondyls').
Ruta et al. (2003) and Ruta & Coates (2007) pro-
posed a monophyletic origin of extant amphibians
among temnospondyls, and placed 'lepospondyls'
among reptiliomorphs (Fig. 1e). According to that
phylogeny, seymouriamorphs and 'lepospondyls'
can be expected to retain the ancestral saltwater tol-
erance, but no inferences can be drawn about habitat
preferences in temnospondyls.
Anderson (2007) proposed a diphyletic origin of
extant amphibians, with batrachians (anurans and
urodeles) nested within temnospondyls, and gymno-
phionans nested among 'lepospondyls', which form
a clade with seymouriamorphs and reptiliomorphs
(Fig. 1f). Under this phylogeny, reptiliomorphs
include only amniotes and diadectomorphs (and
perhaps Solenodonsaurus, which was not included
in the analysis). Under that phylogeny, no infer-
ences can be drawn about the habitat preferences
of
Habitat of early stegocephalians
and their close relatives
A review of the palaeoecological interpretations of
the environment of early stegocephalians reveals
much uncertainty and controversy (Tables 1 & 2).
We have compiled the prevailing interpretations of
the environment of these taxa. When considerable
uncertainty exists, the states which were plausibly
present all appear separated by slashes. However,
interpretations which appear to be significantly
less plausible are not considered, simply because
the amount of uncertainty might be such that little
signal could be extracted.
For example, Tulerpeton was found in a marine
environment located at least 200 km from the
nearest land and, given the completeness and good
preservation of the specimen, it plausibly lived in
the sea (Lebedev & Clack 1993). Nevertheless,
Long & Gordon (2004, p. 704) suggest that the
recovered bones represent a carcass which has
most
early
stegocephalians
(temnospondyls,
