| [Thread Prev] | [Thread Next] | [Thread Index] | [Date Prev] | [Date Next] | [Date Index] |
Nice to see some more comments on this topic.
To say that conodonts probably occupied a similar niche as the fishes
is something of an oversimplification. In the first place, there was no
single "fish niche" during the Devonian, any more than there is today (see
below). Secondly, although I have no problem with the vertebrate affinities
of conodonts, they were in virtually every ecologically important respect
(size, shape, head morphology, environmental preference etc.) quite
different from the Old Red Sandstone fishes I work with. So while the
conodont comparison is interesting, I'm not sure that it tells us that much
about what the fishes were up to.
Some more details about the Devonian vertebrate record:
The Devonian has yielded a lot of vertebrate remains. Carroll's list is
nowhere near comperhensive, and leaves out several important groups
(placoderms, acanthodians, chondrichthyans, osteostracans,
heterostracans) altogether. The most thoroughly studied areas, such as
Scotland and the Baltic States, have produced fish assemblages which
can probably be compared with the dinosaur faunas of western North
America in terms of diversity and stratigraphic resolution. Still
nothing to compare with the marine microfossil record of course,
but pretty good as vertebrates go.
Several patterns emerge from this record. In the first place there is a
strong contrast between an open marine vertebrate record dominated by early
sharks and ray-finned fishes (typically represented only by teeth; look out
for them in your conodont residues!) and a "continental" record dominated
by placoderms, acanthodians, osteostracans, heterostracans, lungfishes,
osteolepiforms and porolepiforms. Some of the "continental" faunas come
from what seem to be fresh-water deposits, but others are definitely
marginal marine; they seem to be dominated by the same taxa irrespective of
the precise environmental setting (this however needs further work).
Within the "continental" record, there is very dramatic change
during the course of the Devonian. At the beginning of the period,
jawless vertebrates (heterostracans and osteostracans) dominate the
faunas together with placoderms and acanthodians. By the end of the
Devonian the heterostracans and osteostracans have disappeared altogether,
and the placoderms and acanthodians are in decline, whereas the
osteichthyans (osteolepiforms, porolepiforms, lungfishes,
coelacanths, ray-finned fishes) which were very rare indeed in the Lower
Devonian, have become abundant and diverse. The tetrapods also arise from
among the osteichthyans before the end of the Devonian. From the end of
the Devonian to the present day osteichthyans have dominated the
continental fish faunas of the world.
Interestingly, the spread of osteichthyans during the Devonian is
NOT coupled to the origin and early radiation of the group. It is clear
from phylogenetic analyses, and the character complements of the
earliest known osteichthyans, that the initial radiation (separation of
the ray-fins and lobe-fins, evolution of the principal lobe-fin groups)
took place before the beginning of the Devonian. What we are seeing must
be a biogeographical effect. So far almost all the information has come
from Laurussia, which has severely restricted the understanding of
Devonian vertebrate biogeography, but recent work in south China and
Australia has revealed very distinctive local faunas. "Palaeozoic
vertebrate biostratigraphy and biogeography" (ed. J.A. Long: Belhaven
Press, 1993) gives a good snapshot of this developing field.
This is the key part: I am interested specifically in the Upper
Devonian (Frasnian + Famennian) "continental" vertebrate record. The
time interval here is only about 14 million years, and the taxic
diversity is rather greater than John assumes; many of the genera are
polyspecific, and in a few cases (notably Bothriolepis; see Young G.C.
1984, Proc. Linn. Soc. New South Wales 107, 443-473) they have quite well
resolved internal phylogenies.
On the face of it, this record seems to show no abrupt change at the
Fra-Fam boundary, but there is a major and apparently abrupt faunal
transformation at the Dev-Carb boundary. Two questions thus pose
themselves:
1) Is there really no sign of an "event" at the Fra-Fam boundary?
2) Is the apparent end-Devonian extinction really abrupt?
Presumably the appropriate null hypothesis in each case is that no
abrupt "event" has occurred. On this basis I would guess (and I mean guess)
that the null hypothesis would be unfalsified in the first case, and
probably also in the second. There certainly seems to be something
interesting going on at the end of the Devonian, but problems with
sample sizes and stratigraphic correlations are likely to scupper a
proper analysis of that one as yet.
I hope this rather long missive gives a clearer picture of the
problems surrounding Devonian vertebrate history. There is a good deal of
work going on in this area (this week's "Nature" has the latest on Devonian
tetrapods), so both the stratigraphic and phylogenetic/taxonomic
resolution should improve in the long run. The available evidence
apparently fails to show a Fra-Fam boundary extinction event in the
"continental" fish faunas. The putative Dev-Carb boundary extinction is
probably beyound the reach of rigorous analysis for the present, but should
be a target for future research.
Per Erik Ahlberg
Senior Research Fellow
Department of Palaeontology
The Natural History Museum
Cromwell Road, London SW7 5BD, UK
E-MAIL ADDRESS: pea@nhm.ac.uk
Partial index: