Epidendrosaurus: tree-climbing theropod from China

[bh480@scn.org]

From: Ben Creisler bh480@scn.org
The online version of Naturwissenschaften posted a new 
article (not yet in print so the name is technically a 
nomen nudum). Here is the abstract and key passages:

Zhang Fucheng , Zhonghe Zhou, Xing Xu and Xiaolin Wang 
2002. A juvenile coelurosaurian theropod from China 
indicates arboreal habits. Naturwissenschaften online 8-21-
2002
Abstract. Here we report an unequivocal arboreal 
coelurosaur, Epidendrosaurus ningchengensis gen. et sp. 
nov. This juvenile coelurosaur's third manual digit is 
extremely elongated, distinctively different from that of 
other known dinosaurs and birds. It represents certainly a 
type of adaptation previously unreported from the Mesozoic 
although the exact function of the third manual digit is 
unclear. The relatively long forelimb, penultimate phalanx 
of manual digit II, and pedal penultimate phalanges, are 
interpreted as evidence for the arboreal habit of 
Epidendrosaurus. Because Epidendrosaurus is more similar 
to advanced birds in some arboreal features than to 
Archaeopteryx, we suggest that the initial appearance of 
tree-adaptation in theropods was probably not directly 
related to flight but to other functions, such as seeking 
food or escaping from predators. 
..
Description of specimen
The specimen is about the size of a house sparrow (Passer 
domesticus) and obviously a juvenile individual as 
evidenced by many distinctive juvenile features, such as 
the incomplete ossifications of articular surface, 
presence of grooves on some limb bones and less well-
defined extremities of postcranial long bones...
Taxonomy
Theropod Marsh, 1881
Coelurosauria Huene, 1914
Maniraptora Gauthier, 1986
Epidendrosaurus ningchengensis gen. et sp. nov.
Etymology
The generic name is derived from the obvious arboreal 
adaptation of this animal, the specific name from the 
locality Ningcheng County of Nei Mongol.
Holotype
IVPP (Institute of Vertebrate Paleontology and 
Paleoanthropology, Beijing, China) collection number 
V12653.
Locality and horizon
Daohugou, Ningcheng County, East of Nei Mongol. Daohugou 
Formation, (?) Late Jurassic (Zhang 2002 )
Diagnosis
Manual digit III elongated, nearly twice as long as digit 
II. Metacarpals II and III short, about 30% of the length 
of humerus. Second phalanx of manual digit II long, nearly 
170% the length of the first phalanx (see electronic 
supplementary material S1).
Description
The frontal and parietal are similar to those of 
dromaeosaurs, such as Sinornithosaurus (Xu et al. 1999 ), 
and the basal bird Archaeopteryx (Elzanowski and 
Wellnhofer 1996 ; Martin and Zhou 1997 . The frontals are 
probably only slightly fused near the caudal region . 
However, the two parietals appear to be completely fused 
as in Sinornithosaurus . There are at least 12 teeth in 
the mandible. The teeth progressively decrease in size 
caudally and are sparsely distributed as in most early 
birds including Archaeopteryx. The jaws are wide. The two 
mandibles are tightly contacted at the anterior end, yet 
it is difficult to determine whether they are fused. Among 
non-avian theropods, only oviraptorids have fused 
mandibular symphysis.
There are at least nine cervical vertebrae preserved in 
articulation, with the anterior ones longer than the 
posterior. The total length of the tail is estimated to be 
6-7 times as long as the femur .Near the distal end of the 
tail, integumentary imprints are faintly preserved, 
resembling those of Microraptor (Xu et al. 2000 ).
The scapula has a primitive expanded distal end (Currie 
and Padian 1997 ). The coracoid is generally similar to 
that of some advanced non-avian coelurosaurs (Currie and 
Padian 1997) and Archaeopteryx (Ostrom 1976 )>; Wellnhofer 
1992 ). The ulna is posteriorly bowed and slightly wider 
than the radius. The manus is considerably longer than the 
humerus. Metacarpals II and III are comparatively short 
and only about one-third the length of the humerus. The 
first phalanx of digit II is shorter than the second, as 
in some other coelurosaurian dinosaurs (Currie and Padian 
1997 ) and the basal birds Archaeopteryx ..., 
Confuciusornis ...and Protopteryx ; the reverse is true in 
more advanced birds such as Sinornis (Sereno and Rao 1992)
and Cathayornis (Zhou 1995 ). One of the most distinctive 
features of the hand of Epidendrosaurus is the extremely 
elongated digit III; it is nearly twice the length of 
digit II, differing from all other known dinosaurs and 
birds (Currie and Padian 1997) .
The hindlimb is slightly shorter than the forelimb mainly 
because of the extremely elongated manual digit III. The 
femur is shorter than the humerus. Metatarsal IV is 
slightly shorter than both II and III. Metatarsal I is 
attached to the distal end of metatarsal II. Distally the 
trochlea of metatarsal I aligns with those of II and III 
as in advanced perching birds, but not in other known 
dinosaurs. The pedal digits are relatively short. Among 
the four pedal digits, IV is longer than II and closer to 
III in length. The penultimate phalanges of digits III and 
IV are markedly longer than the neighboring proximal 
phalanges, whereas in digit II the second phalanx is only 
slightly longer than the first. Digit I is relatively 
long, and distally it reaches the middle of the second 
phalanx of digit III .
Discussion
Many of the features of Epidendrosaurus such as the 
structures of the foot, the hand and the frontal with a 
deep cerebral fossa clearly show that it is a coelurosaur. 
Epidendrosaurus has also preserved several characters such 
as the long forelimb compared to the hindlimb, ulna bowed 
posteriorly, and pedal digit IV longer than II and closer 
to III in length, indicating that it probably belongs to 
the Maniraptora. Phylogenetic analysis has shown that 
Epidendrosaurus is very close to the transition to birds 
(see electronic supplementary material: S2, S3). Due to 
the incomplete preservation of a juvenile individual, many 
of the diagnostic features are less certain than in an 
adult specimen. Therefore, its phylogenetic position as 
suggested in this paper is at best tentative.
Epidendrosaurus is distinguishable from all other 
coelurosaurs in having an extremely elongated third manual 
digit. Manual digit III is nearly twice the length of 
manual digit II. In both juvenile and adult non-avian 
theropods such as Sinornithosaurus, Deinonychus, and the 
basal bird Archaeopteryx manual digit III is shorter than 
II. In the basal dinosaur Herrerasaurus manual digit III 
is only slightly longer than II . The extremely elongated 
manual digit III does not appear to be well adapted for 
grasping; however, it probably cannot be completely 
excluded from playing a minor role in the grasping 
activity of the hand. One possible explanation for the 
elongated third manual digit is that it is adapted to a 
specific niche as in the Malagasy primate Aye-Aye 
(Daubentonia madagascariensis), which uses its long middle 
finger as a tool for finding insects in trees.
The phalangeal proportions of the manus are also 
noteworthy. In digit II, the second phalanx is 
significantly elongated and about 170% as long as the 
first, more than those of other theropod dinosaurs and 
basal birds. The forelimb is slightly longer than the 
hindlimb. In other non-avian theropods, such as 
dromaeosaurids, oviraptorids, troodontids and 
Protarchaeopteryx, the forelimb is shorter than the 
hindlimb. ..... These features, especially the long second 
phalanx of manual digit II and the forelimb, probably 
indicate grasping (Benton 1997 ) or grabbing capability.
A relatively long manus is better adapted to grasping or 
grabbing than to flapping flight; therefore in the early 
evolution of birds, the manus became progressively reduced 
and shortened (Zhou 1995 ). It has been suggested that 
during the early evolution of the tree-adaptation of birds 
the forelimb had played a significant role, i.e., the long 
forelimbs and the long and curved claws were important 
tools in assisting the arboreal life of early birds .... 
Only in later advanced birds did the role of the forelimb 
in arboreal adaptation became gradually reduced. The 
discovery of the new arboreal coelurosaur with long 
forelimbs and penultimate phalanges appears to confirm 
this hypothesis. The forelimb of Epidendrosaurus provides 
further evidence that the climbing/grasping ability is not 
limited to the hindlimb ...The elongation of the forelimb 
is related to arboreal adaptation rather than directly to 
flight. The discovery also supports the hypothesis that 
the forelimb played an important role in the arboreal life 
of coelurosaurs, and lends credence to the arboreal 
hypothesis of the origin of avian flight ...
The foot of Epidendrosaurus is unique among non-avian 
theropods. Although it does not preserve a reversed 
hallux, metatarsal I is articulated with metatarsal II at 
such a low position that the trochleae of metatarsals I-IV 
are almost on the same level , which is similar to those 
of perching birds including the Early Cretaceous flying 
birds Sinornis (Sereno 1992 )and Longipteryx (Zhang and 
Zhou 2001), as well as many arboreal pterosaurs. This 
strongly suggests that Epidendrosaurus had arboreal 
capability (see electronic supplementary material S4), 
which is further evidenced by the toe proportions, i.e., 
the penultimate phalanges are longer than their 
neighboring proximal phalanges ....Pedal digit I is also 
long; distally it reaches the midpoint of the second 
phalanx of digit II. The pedal unguals are long and 
curved, and are generally comparable to those of 
Microraptor and Archaeopteryx ...The dromaeosaur 
Microraptor (Xu et al. 2000 )is the only other known 
theropod that has possessed arboreal capability, although 
the presence of arboreal dromaeosaurs has been proposed 
earlier (Chatterjee 1997 >). However, the relative 
position of the articulation of metatarsal I to II is 
higher than that in Epidendrosaurus. The articulation 
between metatarsals I and II in Epidendrosaurus is even 
lower than that in the basal arboreal birds Archaeopteryx 
and Confuciusornis, which possessed perching ability ...It 
should also be mentioned that since the holotype of 
Epidendrosaurus is a juvenile individual, the proportions 
of the postcranial skeletal elements will certainly be 
somewhat different from an adult. Trees could provide 
relatively safe niches for the defenseless nestlings and 
juveniles.
It seems that the climbing function in Epidendrosaurus was 
acquired before birds (Archaeopteryx and other more 
advanced birds). Although the arboreal life in the 
immediate ancestors of birds was critical for the 
development of the flight of birds, the arboreal habit in 
Epidendrosaurus was most likely for food or for escaping 
from predators (Bock 1986 ). Thus the evolution of the 
arboreal capability in theropods was not necessarily 
directly related to flight in the beginning.