Postilla

Le onsen paseetr ite’ pttes ts Sn ae ea PO Se Sas
Septet ees Ae pes arteg ern hate Cmemaet
out Fe RTS Be nearer betes .
BO ey ein cy ah n EDI Dat dl bey LARS NAS

Pear viat Stina genheeda ory ah

Srantyendne AR RET TES UCU RT hates EEOC
Pe hag AL NINE

Pete te

greene

ANG AEE NEIL

Selena

Renee 0 PN SOL I
Pe er res tae

Ee Se
5 a AR et aes

alte Ff a Ra Bees
Ee dk al
er State wht

a ee

MUS. COMP. ZOOL
LIBRARY

S MAR 3 1967
Ve Ost Mla ee

PEABODY MUSEUM OF NATURAL HISTORY

YALE UNIVERSITY
NEW HAVEN, CONNECTICUT, U.S.A.

Number 106 February 13, 1967

A NEW GENUS AND SPECIES OF MARINE DIPNOAN
FISH, FROM THE UPPER DEVONIAN OF CANADA

KEITH STEWART THOMSON
DEPARTMENT OF BIOLOGY AND
PEABODY MUSEUM OF NATURAL History,
YALE UNIVERSITY

INTRODUCTION

Our understanding of the evolutionary history of the lungfishes
(Osteichthyes, Dipnoi) is based almost exclusively upon fossil
material of freshwater forms. Indeed, everything we know about
the fossil and living Dipnoi suggests that this group mostly has a
freshwater habitat. It is of the greatest interest, therefore, to learn
that the two earliest, and most primitive, lungfish species that we
know of, Dipnorhynchus sussmilchi (Etheridge) and D. lehmani
Westoll, were fully marine forms (see Westoll, 1949 and especially
Campbell, 1965). The discovery and study of further marine lung-
fishes has thus become a matter of some importance, both in
elucidating the variously primitive and advanced characters of
Dipnorhynchus and in establishing more fully the evolutionary
history of the Dipnoi as a whole.

The subject of this short paper is a specimen of lungfish from
the marine Upper Devonian of Canada. The fish clearly belongs
in a new genus and species, and seems to be more primitive than
the typical and well-known freshwater dipnoans of the same age.

to

Postilla Yale Peabody Museum No. 106

DESCRIPTION
Family uncertain
Sunwapta genus novum
grandiceps species novum

‘“Coelacanth remains” (Gardiner 1966, p. 93)

DiaGnosis. A very large dipnoan fish: estimated length of skull
approximately 22 cm., estimated standard length approximately
1 metre. Marine in occurrence. Tooth plates large and lacking
separate denticles. Lower tooth plates differentiated peripherally
by the presence of grooves separating approximately parallel ridges
that show faint signs of becoming subdivided into small denticles.
Dorsal surfaces of the ridges covered with enamel similar to that
covering the dermal bones. Central and posterior portions of the
tooth plates smooth and lacking the enamel covering (in all prob-
ability this is lost through wear during the life of the animal).
Dermal bones of the lower jaw massive and bearing an external
covering of enamel identical with that found in typical Dipterus
Sedgwick and Murchison.

OCCURRENCE. Sunwapta Pass, Mount Athabasca, Alberta, Canada.
Upper Devonian. Precise locality unknown.

TyPE SPECIMEN. Specimen number 477, University of Alberta
Geological Museum, Edmonton, Alberta. Collected by “Mr. M.
Bleuler,” date unknown.

REMARKS. Unfortunately only a single specimen of this fish is
known. The specimen is well preserved (Fig. 1) and consists of
the symphysis and the anterior portions of the rami of the lower
jaws. The specimen shows no sign of pronounced weathering or
rolling and there can be virtually no doubt but that it derives from
the marine environment in which it was deposited rather than
having been transported before or after fossilization from some
distant freshwater locality. The nature of the tooth plates also
tends to confirm the conclusion that the fish lived in a marine
environment, the structure suggesting that the fish was adapted to
feeding on molluscs or other hard-shelled invertebrates. It has

FIGURE 1. Sunwapta grandiceps gen. et sp. nov. Holotype,
symphysial region of the lower jaws in occlusal
and anterior view. Natural size.

1967 A New Devonian Marine Dipnoan 3

been difficult to assess the familial status of the fish. The general
structure shows a clear affinity with the families Dipteridae and
Dipnorhynchidae, but the fish probably does not correctly belong
in either of these families. The problem is discussed further below.

DISCUSSION

A search through the literature pertaining to the Dipnoi reveals
that fragmentary remains of large lungfish have been found in a
wide variety of Devonian marine or semi-marine deposits. Among
such forms may be included Palaeadaphus van Beneden and
Koninck, possibly Grossipterus Obruchey (formerly Holodus),
Conchodus McCoy, Holodipterus White and Moy-Thomas (for-
merly Archaeotylus) in addition to Dipnorhynchus sussmilchi
and D. lehmani. These fishes are mostly imperfectly known; in the
majority of cases only portions of the mandibles and tooth plates
have been preserved. Nonetheless, we may see that they all have
in common a marine origin, moderately large to very large size,
and tooth plates which lack separate denticles and bear only a small
number of broad, faintly ornamented, ridges radiating from a point
near the postero-medial corner of the plate. Naturally, it is highly
important in this analysis to be sure that the patterns of denticula-
tion (particularly the lack of denticulation) are not simply caused
by wear of the specimen. Usually, as in the present case, this can
be determined quite readily.

It is of considerable interest to speculate upon the phylogenetic
history of these marine forms. Since, in the majority of cases, only
the tooth plates are known to us, it is impossible at the present time
to determine with accuracy the morphological resemblance and
taxonomic relationship of these fishes to the better-known fresh-
water Dipnoi of the Devonian. It seems reasonable to categorize
them as separate genera and species close to but quite distinct
from the family Dipteridae; it is probably most economical to
assign them as Dipnoi incertae sedis.

A more important question than the taxonomic relations of
these marine forms concerns their relative primitiveness. Are they
secondarily derived from freshwater forms similar to the Dipteridae
etc., or do they represent remnants of an ancestral marine radia-
tion of Dipnoi from which all the freshwater forms have later
evolved? At present we are only able to discuss the question with

4 Postilla Yale Peabody Museum No. 106

respect to the evidence afforded from the structure of the tooth
plates.

The studies of White (1965, 1966) have cleared up a good
deal of the confusion concerning the nature of the dentition in
Dipnoi. As he states, the Dipterus-type of dentition “is formed
from specialized areas of the dentine-covered bone surface (of the
jaws) and... the denticles arise subsequently to the development
of the crushing plate” (1966, p. 7). According to this view the
relatively unspecialized dentition of Dipnorhynchus (White, 1964;
Campbell, 1965) is strictly primitive and has not been preceded
by any denticulated stage.

Since, as the evidence stands at present, it is difficult to inter-
pret the nature of the tooth plates in Dipnorhynchus as other than
primitive, the characteristic denticulation of the tooth plates of the
freshwater Dipnoi such as Dipterus, Fleurantia Graham-Smith and
Westoil, Scaumenacia Traquair, or Rhinodipterus Save-Soderbergh
must be interpreted as having arisen secondarily, probably in
accordance with adaptation to a new feeding requirement in a new
habitat. These freshwater fishes seem to be adapted for dealing
with a less brittle food; the difference between feeding upon
arthropods and upon molluscs. (No doubt, of course, Dipterus
and its relatives also fed on smaller freshwater shellfish and the
marine forms probably fed on larger crustaceans too.) It is clear,
however, that the denticulation of the Dipterus-type of tooth
allows for the fish to deal with a wider range of food than the
simple crushing plates of Dipnorhynchus. In a previous study
(Thomson, 1965) I have discussed the possibilities that exist for
differentiation of the dental battery in various Dipnoi — for exam-
ple, by the development of a central crushing area quite dis-
tinct from the peripheral denticulation which may be used for
“chopping.”

If this interpretation is correct then Sunwapta and the similar
large marine forms that we have listed above most probably belong
to a group intermediate between the early Dipnorhynchidae and
the later Dipteridae. They may represent part of an early radiation
of Dipnoi from which the freshwater forms later evolved. The
simplest interpretation of the pattern of the tooth plates in the
forms like Sunwapta or Palaeadaphus is that they represent the
earliest stage towards the development of the Dipterus-type of

1967 A New Devonian Marine Dipnoan 5)

tooth plate. According to this hypothesis, therefore, in the phyl-
ogenetic history of the Dipnoi the initial stage of the tooth plate
from the coarsely tubercular Dipnorhynchus-stage is the differentia-
tion of a series of marginal ridges, formed by the excavation of
deep grooves in the peripheral portion of the plate. Thus the
ridges are not added to the plate but are formed by the develop-
ment of the grooves. The curvature and location of the opposing
plates apparently allows the upper and lower ridges to interdigitate.
We may extend this hypothesis to suggest that the formation of
discrete denticles on the tooth plate occurs through subdivision of
the ridges on the tooth plate. This hypothesis would accord well
with White’s observations on the nature of the denticles in Dipnoi
since the denticles would essentially represent the original enamel-
covered surface of the tooth plate.

In view of the marine affinities of Dipnorhynchus itself (the
earliest dipnoan that is currently known), the alternative view-
point — that the tooth plates in these fishes represent a secondary
adaptation to marine life that has evolved from an ancestral fresh-
water dipterid stock — is less satisfactory. We may note, however,
that the marine (or estuarine) Dipterus digitatus Eastman, D.
mordax Eastman, D. pectinatus Eastman and D. costatus Eastman
from the Upper Devonian of Iowa (Eastman, 1908) most probably
do have just such a derivation from more advanced freshwater
forms. Their exact taxonomic position is uncertain and will not be
discussed here. They are clearly different from the Sunwapta —
Palaeadaphus type in that, although the main body of the tooth
plate is frequently not denticulated, distinct denticles are present
nearer the periphery. Another feature that may be important is
that in these apparently secondary forms the tooth ridges “radiate”
from a point in the mid-section of the tooth plate, whereas in the
more primitive forms the ridges are more parallel to the antero-
posterior axis. On the whole, it seems most likely that these forms
have arisen from more typical advanced dipnoans by the loss of
denticles from the median portion of the tooth plate. Again, this
development of a crushing type of dentition is associated with a
marine or estuarine environment.

6 Postilla Yale Peabody Museum No. 106

ACKNOWLEDGMENTS

I am grateful to Dr. C. Stelck and the Department of Geology,
University of Alberta, Edmonton, for the loan of the specimen
described here. The study was supported financially by National
Science Foundation grant number GB 4814.

LITERATURE CITED

Campbell, K. S. W. 1965. An almost complete skull roof and plate of the
dipnoan Dipnorhynchus sussmilchi (Etheridge). Palaeontology 8:
634-637.

Eastman, C. R. 1908. Devonian fishes of Iowa. Iowa Geol. Surv. 18:
29-386.

Gardiner, B. G. 1966. Catalogue of Canadian fossil fishes. R. Ont. Mus.,
Contrib. 68: 1-154.

Thomson, K. S. 1965. On the relationships of certain Carboniferous Dipnoi;
with descriptions of four new forms. Roy. Soc. Edinburgh, Proc., B.
69: 221-245.

Westoll, T. S. 1949. On the evolution of the Dipnoi. p. 121-184. Jn G. L.
Jepsen, E. Mayr, and G. G. Simpson, [eds.]. Genetics, paleontology
and evolution. Princeton Univ. Press, Princeton.

White, E. I. 1965. The head of Dipterus valenciennesi Sedgwick and Murchi-
son. Brit. Mus. Nat. Hist., Geology Bull. 2: 1-45.

1966. Presidential address: A little on lung-fishes. Linn.
Soc. London, Proc. 177: 1-10.

=

3 PF

ss te eet
Fir PE f
SO re ALES

Be dao
era

ears