taxonomic status of edaphosaurus raymondi case

TAXONOMIC NOTES
1049
'based on Astropecten pewei Miller and Unklesbay from the
Upper Jurassic Sundance Formation of Wyoming) has a rela­
tively smaller disc, with more angular inter-ray margins and
smaller actinal (intermediate) fields whose plates do not extend
rnto the arms.
GRAY, J. E. 1840. A synopsis of the genera and species of the class
Hypostoma (Asterias Linn.). Annals aud Magazine of Natural His­
tory, 6:175-184,275-290.
.
HESS, H. 1955. Die fossilen Astropectinideu (Astewidea). Schweiz­
_.
erische Palaeontologische Abhandlungen, 71:1-113.
- - . 1960a. Ueber die Abgrenzung der Astropectiniden-Gattungen
Pentasteria Valette und Archastropecten Hess. Ec10gae geologicac Hel­
ACKNOWLEDGMENTS
vetiae, 53:329-331.
The specimen described here was found and made available - - . 1960b. Pentasteria (ArchasLropecten) procera n. sp. (Asteroidea,
Astropectinidae) ans dem Bajocien von Cheltenham (England). Eclo­
for study by Suzan Moore. We extend sincere thanks to D. B­
gae geologicae Helvetiae, 53:331-334.
lake for his invaluable assistance in helping us recognize certain
ssicles and their significance for taxonomic placement ofthis - - . 1968. Ein neuer Seestem (Pentasteria longispina n. sp.) aus den
Effingerschichten des Weissensteins (Kt. Solothum). Ecologae geo­
=~Ispecimen. Recognition of the ossicle series preserved in this
logicae Helvetiae, 61:607-614.
" specimen was also greatly facilitated by comparison with the - - . 1973. Neue Echinodennenfunde aus dem mittleren Dogger des
excellent analysis and illustrations of test elements in living
Aargauer Juras. Eclogae geologicae Helvetiae, 66:625-656.
members of the Order Paxillosida provided by Hess (1955) and - - . 1987. Neue Seestemfunde aus dem Dogger des Schweizer Juras.
Eclogae geologicae Helvetiae, 80:907-918.
Blake (1981, 1987). Rosemary Neuman drafted the diagrams
McLEARN, F. H. 1944. The occurrence of starfish in the Lower Cre­
and Rick Larush produced the photographic illustrations.
taceous of the Peace River Valley. The Canadian Field-Naturalist,
58:132-134.
REFERENCES
PERRIER, J. O. E. 1884. Memoire sur les etoiles de mer receuillies dans
la Mer d'Antilles et Ie Golfdu Mexique. Museum d'Histoire Naturelle,
LAKE, D. B. 1973. Ossiele morphology of some Recent asteroids and
Paris, Nouvelles Archives, series 2, 6: 127-276.
description of some west American fossil asteroids. University of
RASMUSSEN, H. W. 1944. Observations on the Astewid Fauna of the
California Publications in Geological Sciences, 104:1-59.
Danian. Meddelanden fra Dansk Geoliske Forenlingen, 10, Heft 4:
--. 1981. The new Jurassic sea star gen us Eokainaster and com­
417-426.
ments on life habits and thc origins of the modem Asteroidea. Journal
SLAOEN, W. P. 1889. Report on the scientific results of the voyage of
of Paleontology, 55:33-46.
H.M.S. Challenger during the years 1873-76. Zoology, Volume 30,
-. 1986. Some new post-Paleozoic sea stars (AsterQidea: Echinoder­
Report on the Asteroidea. H. M. Stationery Office, London, 893 p.
mata) and comments on tax.on endurance. Journal of Paleontology,
60: 1103-1119.
ACCEPTED 9 MAy 1990
-. 1987. A classification and phylogeny ofpost-Palaeozoic sea stars
(Asteroidea: Echinodennata). Journal of Natural History, 21 :481­
The Natural Science and Engineering Research Council provided $200
in support of this article.
528.
iii
t
1.5,
. Poleo"t.• 64(6),1990. pp. 1049-1051
'opyright © 1990, The Paleontolog,cal SOCIety
'22·3360/9010064-1049$03.00
TAXONOMIC STATUS OF EDAPHOSAURUS RAYMONDI CASE
S. P. MODESTO AND R. R. REISZ
Department of Zoology, Erindale College, University of Toronto,
Mississauga, Ontario LSL lC6, Canada
HE EDAPHOSAURIDAE is an important family of primitive syn­
psids that includes the oldest known terrestrial herbivorous
ertebrates. Until the recent description of a new, primitive
phosaur from the Upper Pennsylvanian of Kansas (Reisz
d Berman, 1986), the herbivorous Edaphosaurus was the only
own member of the family. Accordingly, fragmentary eda­
hosaurid materials from Europe and North America had been
signed to Edaphosaurus on the basis of the characteristic neu­
I spine morphology.
One such specimen was eollected with the remains of several
ther tetrapods at the beginning ofthe century by P. E. Raymond
om an Upper Pennsylvanian locality near Pitcairn, Pennsyl­
-ania (Case, 1908). This small fragment of neural spine was the
Idest edaphosaurid material then known, and the provisional
On Naosaurus raymondi was erected by Case (1908) to cover
orth American species with neural spines of this size.
A quarter century before the discovery of the Pitcairn spec­
en, Cope (1882) had established the genus EdaphosauTus on
e basis of a skull and axis from the Permian of north-central
exas. Subsequently, postcrania with elongated, crossbarred
:ural spines recovered from the same area were attributed to
e genus NaosauTus (Cope, 1886). Case (1907) realized that the
11 of the former belonged with the vertebrae of the latter,
and with Williston (1913) tentatively synonymized Naosaurus
with Edaphosaurus. However, Williston (1918) continued to use
the term Naosaurus, and it was not formally relegated to syn­
onymy until 1940 by Romer and Price.
Case (1907) evidently believed that the small size of the Pit­
cairn specimen merited specific status and did not consider its
possible relationship to Edaphosaurus mirabilis, an equally di­
minutive form from the Upper Pennsylvanian of Kounova,
Czechoslovakia. This may have been because he misinterpreted
the size of Fritsch's (1895) illustration (Romer and Price, 1940).
For lack of adequate comparative material of the same age,
miniature edaphosaurid neural spines collected from the Sangre
de Cristo Formation (Upper Pennsylvanian) ofcentral Colorado
were referred by Vaughn (I969) to Edaphosaurus alf. E. ray­
mondi.
Romer and Price (1940) retained the specific status of Eda­
phosauTUs raymondi, acknowledging that it was separable from
E. mirabilis only on geographical grounds. Reisz (1986) ques­
tioned the validity of E. raymondi as a distinct species because
the holotype is diagnostic only at the familial level. For the same
reason, the validity of the Kuonova species, known only from
a single vertebra, has also been questioned (Reisz, 1986).
Descriptions ofthe small, primitive edaphosaur Ianthasaurus
1050
JOURNAL OF PALEONTOLOGY, V. 64. NO.6, 1990
FIGURE 1- Edaphosaurus raymondi, CM 1941, spine fragment. 1, an­
terior? and posterior? views; 2. lateral views. Scale = 5 mm.
hardestii from the Upper Pennsylvanian of Kansas (Reisz and
Berman, 1986; Modesto and Reisz, 1990) have necessitated
reexamination of neural spine material of small size attributed
to Edaphosaurus. Ianthasaurus, represented by several speci­
mens, is the oldest known edaphosaur and its neural spines are
virtually indistingushable, except for their smaller size, from
those of Edaphosaurus. These recent increases in our knowledge
of the North American edaphosaur assemblage permit a thor­
ough comparison of Edaphosaurus raymondi with other eda­
phosaurs. This reevaluation is an attempt to determine whether
the Pitcairn specimen is referable to either Edaphosaurus or
Ianthasaurus, or if E. raymondi is actually an invalid taxon.
SYSTEMATIC PALEONTOLOGY
Subclass SYNAPSIDA Osborn, 1903
Order EUPELYCOSAURIA Kemp, 1982
Family EDAPHOSAURIDAE Cope, 1882
Genus EDAPHOSAURUS Cope, 1882
EDAPHOSAURUS RAYMONDI Case, 1908
Figure 1
EdaphosauTUs raymond; CASE, 1908, p. 237-238, fig. 7, PI. 59, fig. 3;
ROMERANDI'RrCE, 1940, p. 19, p. 388; ROMER., 1952,p. 93; PEABODY,
1957, p. 948; VAUGHN, 1969, p. 24; BEIlMAN, 1979, p. 186; REISZ,
1986, p. 72; REISZ AND BERMAN, 1986, p. 89; MODESTO AND REISZ,
1990, p. 834-844.
Material. -Carneigie Museum of Natural History specimen
CM 1941 (Figure 1). This fragment is the sole evidence ofeda­
phosaurids in the Pittsburgh red shale (Conemaugh Group, Up­
per Pennsylvanian) at Pitcairn, Pennsylvania.
Description. -Shaft straight, 12 mm long, oval in cross sec­
tion, with long axis aligned anteroposteriorly; anterior and pos­
terior sides cannot be distinguished from one another and de­
void of longitudinal ridges or grooves; mere stumps remain of
lateral tubercles; one pair of lateral tubercles lies at one end of
fragment with second pair just short of the other; all tubercle
stumps appear to be directed laterally and slightly towards same
end of fragment; one pair of lateral tubercles lies directly op­
posite one another, remaining pair staggered; whether fragment
belonged to mature or immature animal is not determinable.
Discussion. -Case's (1908) suggestion that the fragment is
from near the arch because one end is transversely wide is ten­
uous, The end appears large because the break includes the base
of one of the lateral tubercles. However, Case (1908) was prob­
ably correct in his identification of the proximal and distal ends
of the fragment, since lateral tubercles are more commonly di­
rected slightly dorsally than ventrally in specimens of both eda­
phosaur genera. The absence of either a longitudinal groove or
ridge is not useful in determining affinities to either genus with­
out knowing the location of the fragment on the neural spine.
All neural spines of Ianthasaurus lack longitudinal ridges distal
to the basal lateral tubercles, whereas longitudinal grooves and
ridges originating proximally on the neural spines vary in length
in Edaphosaurus. Accordingly, the morphological evidence does
not bar the Pitcairn fragment from belonging to either a mature
Ianthasaurus or an immature Edaphosaurus.
Both Edaphosaurus and Ianthasaurus are present in strata of
Pennsylvanian age of North America. Ianthasaurus hardestii is
well represented by both mature and immature specimens from
the uppermost Missourian of Kansas. Neural spine materi
attributed to the medium-sized E. novomexicanus has been col­
lected from Upper Pennsylvanian strata in El Cobre Canyon.
New Mexico. Edaphosaurus colohistion, a medium-sized eda­
phosaur from West Virginia, comes from deposits that have
been recognized as either Upper Pennsylvanian or Lower Penn­
ian (Berman, 1979). The validity of E. mirabilis has been ques­
tioned by Reisz (1986). Modesto and Reisz (1990) have sug­
gested that, by virtue of its early occurrence and small size, the
Kuonova edaphosaur may be closely related to the insectivorous
Ianthasaurus hardestii. The stratigraphic position of the Pitcairn
fragment, intermediate to those cited above, is therefore of du­
bious use in associating the Pitcairn fragment with either genus.
In the absence of diagnostic features below the familial level
in the Pitcairn specimen, it is therefore necessary to declare
Edaphosaurus raymondi a nomen vanum. The Sangre de Cristo
material, thought by Vaughn (1969) to show close affinities to
E. raymondi, should be assigned to Edaphosauridae incertae
sedis. Nevertheless, the Pitcairn fragment remains an important
specimen attesting to the widespread success of this group 0:'
pelycosaurs in the Pennsylvanian.
ACKNOWLEDGMENTS
We wish to thank David S. Berman for the loan of the spec­
imen and for critically reviewing the manuscript. We are al
indebted to Heather M. Wilson and Don B. Brinkman for com,
mentary, whieh has improved this paper.
REFERENCES
BERMAN, D. S. 1979. Edaphosaurus (Reptilia, Pelycosauria) from tht
Lower Permian of northeastern United States, with description ofz
new species. Annals of the Carnegie Museum, 48: 185-202.
CASE, E. C. 1907. A revision of the Pelycosauria of North America.
Carnegie Institution of Washington, Publication 55: 1-176.
- - . 1908. Description of vertebrate fossils from the vicinity of Pitts­
burgh, Pennsylvania. Annals of the Carnegie Museum, 4:234-241.
COPE, E. D. 1882. Third contribution to the history of the Vertebra12.
of the Permian formation of Texas. Proceedings of the Arnericar.
Philisophieal Society, 20:447-461.
- - . 1886. The long-spined Theromorpha of the Permian epoch.
American Naturalist, 20:544-545.
FRITSCH A. 1895. Ueber neue Wirbelthiere aus der Permiformatio:.
Bohm~ns, nebst einer Uebersicht der aus derselben bekannt gewor·
denen Arten. Sitzungsberichte der Bohmischen Gesellschaft der Wis,·
senschaften, 53:1-17.
KEMP, T. S. 1982. Mammal-like Reptiles and the Origin ofMammal!
Aeademic Press, London, 363 p.
MODESTO, S. P., AND R. R. REISZ. 1990. A new skeleton of lanthasa'
rus hardeslii. a primitive edaphosaur (Synapsida: Pelycosauria) fro­
the Upper Pennsylvanian of Kansas. Canadian Journal of Earth Se­
ences, 27:834-844.
OSBORN, H. F. 1903. On the primary division of the Reptilia into t....
subclasses, Synapsida and Diapsida. Science, 17:275-276.
PEABODY, F. E. 1957. Pennsylvanian reptiIcs of Gamett, Kansas: edz­
phosaurs. Journal of Paleontology, 31:947-949.
REISZ, R. R. 1986. Pelycosauria, p. 1-102. In P. WeIlnhofer (ed.
Handbuch der Paliioherpetologie, Part 17A. Gustav Fischer Verl36­
Stuttgart.
--.AN
itive
vanian
Earthl
ROMER,
verte
Carneg
- - , ANI
Socie
VAUGHN
.T. Paleo
Copynglll
0022-33
TAXONOMIC NOTES
D. S. BERMAN. 1986. Ianthasaurus hardestii n. sp., prim­
itive edaphosaur (Reptilia, Pelycosauria) from the Upper Pennsyl­
vanian Rock Lake Shale near Garnett, Kansas. Canadian Journal of
Earth Sciences, 23;77-9l.
ROMER, A. S. 1952. Late Pennsylvanian vertebrates and Early Penn ian
vertebrates of the Pittsburgh-West Virginia region. Annals of the
Carnegie Museum, 33:47-113.
- - , AND L. I. PRICE. 1940. Review of the Pclycosauria. Geological
Society of America Special Paper 28, 538 p.
VAUGHN, P. P. 1969. Upper Pennsylvanian vertebrates from the San-
- - . AND
1051
gre de Cristo Formation of Central Colorado. Los Angeles County
Museum Contributions in Science, 164: 1-28.
WIllISTON, S. W. 1918. The osteology of some American Permian
vertebrates, III. Contributions from Walker Museum, 2;87-112.
- - , AND E. C. CASE. 1913. A description of Edaphosaurus Cope.
Carnegie Institution of Washington, Publication 181:71-81.
ACCEPTED 1 JUNE 1990
The Department of Zoology, University of Toronto provided $75 in
support of this article.
~~. 1051-1054
1990, The Paleontolog>cal Society
0022-3360/90/0064-1051$03.00
J. PaleonI.. 64(6), 1990,
Co~yrighl ©
SYSTEMATIC POSITION OF THE PAULCHOFFATIINAE
(MULTITUBERCULATA, MAMMALIA)
DAVID W. KRAUSE AND GERHARD HAHN
Departments of Anatomical Sciences and Earth and Space Sciences,
State University of New York, Stony Brook 11794, and
Institut flir Geologie und Palaontologie der Philipps-Universiilit,
Universitats-Gebiet Lahnberge, Hans-Meerwein-Strasse, D-3550 Marburg,
Bundesrepublik Deutschland
THE LATE Jurassic Paulchoffatiinae, one of two named subfam­
ilies of the family Paulchoffatiidae (suborder Plagiaulacoidea),
are generally recognized as the most primitive known, undoubt­
ed multituberculates (e.g., Hahn, 1969; Clemens and Kielan­
Jaworowska, 1979; Hahn and Hahn, 1983; Hahn et aI., 1989).
Despite the primitive morphology ofmembers of this subfamily
(Paulchoffatia and Pseudobolodon), three features have been
posited as autapomorphic for the subfamily (or family), and
thus to preclude it from the ancestry oflater multituberculates;
I) the distribution of cusps on M 2 (Hahn, 1969, 1971); 2) the
complex structure of 13 (Hahn, 1969, 1971); and 3) the position
of M2 relative to M' (Van Va1en, 1976; Clemens and Kielan­
Jaworowska, 1979; Hahn, 1987). Van Valen(1976) argued that
the first two features are not unequivocally autapomorphic for
paulchoffatiines and that they could, and indeed should, be
considered primitive for the order. Hahn (1977) responded to
Van Valen's assessment but Clemens and Kielan-Jaworowska
(1979, p. 138) opined that "neither argument is particularly
compelling" in light ofthe poor early record of multituberculates
and the consequent lack ofknowledge concerning morphological
variability.
Despite the recent discoveries ofnumerous specimens ofm u1­
titubercu1atc or multituberculate-like mammals from the Late
Triassic and Early Jurassic (e.g., Jenkins ct aI., 1983; Wouters
et aI., 1984; Hahn et aI., 1987, 1989; Sigogneau-Russell et aI.,
1986; Sigogneau-Russell, 1989), these materials are of isolated
teeth only and, as such, cannot serve to resolve unequivocally
the polarity of states in the first two characters. The condition
of the third purportedly apomorphic feature (the relative po­
sition of M2) heretofore has not been resolved, in part because
there is not a single paulchoffatiine specimen in which M2 is
preserved in situ. Nonetheless, our recent restudy of all of the
relevant materials allows us to demonstrate conclusively that
the position of M2 with respect to Ml in paulchoffatiines is as
in all other known multituberculates and that this character state
is not autapomorphic for paulchoffatiines.
In other plagiau1acoids and, in fact, in all other nonpaul­
choffatiine multitubercu1ates in which this region is known, M2
lies distolingual (rather than directly distal) to M' (Figure 1.1).
Thus, since only two rows of cusps are generally developed in
plagiau1acoids, the buccal row of M2 lies directly distal to the
lingual row ofM'. The buccal row ofMI has no distal continuity
with a row of cusps on M2 and the lingual row of M2 has no
mesial continuity with a row of cusps on MI. In later multitu­
berculates, a third row of cusps is developed distolingually on
M' and mesiobuccally on M2 such that the three rows of cusps
on M2 are, in fact, continuous with three rows of cusps on MI.
Hahn (1969, 1971, 1987), however, claimed that M2 lay directly
distal to M' in paulchoffatiines and that the buccal and lingual
rows on M2 were continuous with those on Ml (Figure 1.2).
The importance of Hahn's observations concerning the rel­
ative position ofM2 were summarized by Clemens and Kielan­
Jaworowska (1979, p. 139):
"Should future discoveries validate Hahn's reconstruction
of the occlusal pattern of the paulchoffatiine dentition, we
would argue that its difference from the occlusal pattern of
all other multitubercu1ates is of great phylogenetic signifi­
cance. First, it would clearly bar paulchoffatiines from the
ancestry of any other multituberculates. Also, it would in­
dicate that the common ancestry ofpaulchoffatiines and other
multitubercu1ates would have been at the evolutionary grade
at which the second rows of cusps were just being added to
the molars. Ifthis is true, a tremendous amount ofparallelism
followed in the evolution of the dentitions of these two lin­
eages .... Further, the differences in molar occlusal pattern
would be so basic and ancient that they would warrant sep­
aration of the paulchoffatiines from all other multitubercu­
lates, possibly at the subordinalleve1."
Van Valen (1976, p. 198), on the basis of examination of
figures of specimen V.I. 110-l55a presented by Hahn (1969,
fig. 9 and PI. 8, fig. 25), was the first to question Hahn's obser­
vations. Van Valen concluded that the alveoli for M2 were in­
deed lingual to those for M', "and to the expected degree." He
further judged that "the relative position does not appear to be
due to distortion, but this is not certain." Clemens and Kielan­
Jaworowska (1979, p. 138) expressed greater doubts in stating:
"Study of the illustrations of this specimen ... suggests to us