Geoscience Reference
In-Depth Information
suggests that they were closely spaced; this distri-
bution is confirmed by the spatial arrangement of
in situ stumps (Banks et al. 1985; Driese et al.
1997). However, these trees that lacked laminar
leaves may not have formed completely closed
canopies (Stein et al. 2007). Allochtonous plant
remains associated with the in situ casts represent
smaller sized plants such as bushy Aneurophytales
and herbaceous lycopsids. It is currently uncertain
whether these remains were components of the
forest understory or if they were washed up into
the fossil deposits during floods or other cata-
strophic events. The Gilboa trees lived in the
Catskill delta, either on levees that were periodically
flooded (Banks et al. 1985; Boyer 1995) or in a wet,
poorly-drained coastal margin setting (Driese et al.
1997). The climate is thought to have been tropical-
dry with possible heavy seasonal rainfalls (Banks
et al. 1985).
Remains of pseudosporochnaleans are abundant
in Laurussia on both sides of the Acadian mountains
(Berry & Fairon-Demaret 2001). They have been
reported to be common in Middle Devonian depos-
its of Siberia and Kazakhstan (Iurina 1988) but
Meyen (1987) questioned the validity of the identi-
fication of fossil plants from these areas. In Gond-
wana, where Devonian plant localities are few and
do not yield well-preserved remains, there are no
records of pseudosporochnaleans apart from the
Wattieza described from the Lower Member of the
Campo Chico Formation in Sierra de Perij´, Vene-
zuela (Berry 2000). It is not certain, however, that
the Sierra de Perij´ terrane was already part of
Gondwana in the Middle Devonian (Berry et al.
2000). Pseudosporochnaleans are unknown in
North China (Cai & Wang 1995) and all previous
records of Pseudosporochnus from South China
have recently been dismissed (Berry & Wang
2006a). The closest genus found in South China is
Rhipidophyton, which is common in Yunnan
where it is represented by axes measuring up to
9 cm in diameter (Berry & Wang 2006b).
(i.e. the only organ able to grow significantly) is
the trunk. The supply of branches that are regularly
replaced may not have changed significantly
during the lifetime of the tree. In the arborescent
pseudosporochnaleans, branches lack leaves. By
contrast, Archaeopteris trees possess leaves and
have long-lived roots and branches (Fig. 5c). Their
growth can be qualified as three-dimensional by
comparison to the essentially vertical growth of
the pseudosporochnalean trees (Algeo et al. 2001;
Meyer-Berthaud & Decombeix 2007).
With the developmental model that we propose
for the arborescent Pseudosporochnales, we chal-
lenge the idea defended by Berry & Fairon-Demaret
(2002) of a lignophytic model of growth for these
trees. We hypothesize that their large size, or more
specifically the large diameter of their trunk above
ground, results from the large size of their primary
body (Fig. 5b). Secondary vascular tissues contrib-
ute little if anything to trunk diameter and support.
The maximal width of the primary body is reached
early during growth. This strategy where the large
size of the trunk is reached early during growth
and does not necessitate the development of second-
ary vascular tissues for mechanical support is very
different from that shown in the lignophyte trees
such as Archaeopteris. The latter invest essentially
in secondary xylem (wood) to achieve diameter
and mechanical support as well as water transport
(Figs 3e & 5d). In the arborescent lignophytes, the
maximal diameter of the trunk is reached late
during growth.
Considering both Algeo's Devonian plant
hypothesis (Algeo et al. 2001) and the Beerling &
Berner (2005) model, the differences assessed in
terms of root system, biomass allocation during
growth and amount of lignous tissues in these two
euphyllophyte strategies suggest that they probably
impacted the environment differently. In addition,
the limited extent of the root system and large
amount of living cells forming the trunk in pseudos-
porochnalean trees suggest that (1) these plants
probably inhabited areas with no water limitation
during the growth season and (2) that they avoided
climates with freezing temperatures. The gymnos-
perm type of vegetative body of archaeopteridaleans
was adapted to a wider range of conditions including
cold climates. The pseudosporochnalean trees were
well represented in a low south latitude warm-
temperate zone in the Middle Devonian (Stein
et al. 2007). Archaeopteridaleans were cosmopoli-
tan in the Late Devonian. The former considerations
may explain these differences
Conclusion
The pseudosporochnalean trees and the Archaeop-
teridales that evolved in the Devonian form two
types of euphyllophyte trees which contrast not
only morphologically but also structurally and
developmentally. Morphologically, recent discov-
eries demonstrate that the pseudosporochnalean
trees resemble tree-ferns, that is, that they possess
a trunk bearing adventitious roots at base and
topped by a narrow crown of deciduous branches
(Stein et al. 2007) (Fig. 5a). Individual roots are a
few centimetres wide and have a limited vertical
extent. Above ground, the only perennial organ
in terms of
palaeogeographical distribution.
We thank L. VanAller Hernick, F. Mannolini, B. Stein and
C. Berry for permission to visit their working collections
of plants from Gilboa. These fossils are housed in the
