148 I THE HABITAT OF THE DINOSAURS
considerable area. In other words, the fingers supporting the web
would have to be long and spread out. Present-day ducks are ef-
fective paddlers—with their hind legs, of course—and their hind
toes are exceptionally long and widespread, so that the paddling
web is relatively huge compared to the bird's body size. Beavers
are among the best of modern mammalian web-propelled pad-
dlers, and their toes feature a similar wide-spreading arrangement.
Otters, muskrats, and bullfrogs follow the same rule. But duckbill
dinosaurs don't. The forepaws of the wide-billed Edmontosaurus or
of the crested Parasaurolophus are the exact opposite of what could
be expected in a specialized paddle. Those duckbills' front toe bones
are short and the three main fingers are carried closely together
with hardly any spread at all to the overall hand pattern. A very
strange arrangement indeed if these dinosaurs were as fond of
swimming as orthodoxy says.
So the duckbill's forepaw was manifestly inadequate for effec-
tive propulsion in water; but the orthodox theory maintains a sec-
ond line of defense, the hind feet. Duckbills concentrated almost
all their limb power in their huge hind legs, which were twice the
thickness of their forelimbs. If evolution wanted to design a swim-
ming duckbill, it would be logical to make the hind paw, not the
forepaw, the main underwater propulsive organ. Therefore the toes
of the hind feet should be long and spreading. But, once again,
they are not. The duckbills' hind toes are among the very shortest
ever evolved in the Dinosauria—much shorter, for example, than
those of the meat-eating tyrannosaurs or the plant-eating horned
dinosaurs.
Duckbills trace their lineage back to a gazelle-dinosaur, a small,
long-legged, fast runner like Dryosaurus from Como. Dryosaurus
possessed relatively longer toes than its duckbill descendants and
thus might have paddled better. But according to the orthodox
theory, Dryosaurus was strictly terrestrial in its habits, a confirmed
landlubber, while the shorter-toed duckbills were supposedly
committed to a watery life style. This paradox demands further
scrutiny. Evolutionary processes are supposed to alter adaptations
so that their possessors become better fitted to their new environ-
ments. But duckbills became progressively shorter-toed and
therefore progressively worse adapted for paddling, the very habit
they were supposedly evolving toward.
The argument for the swimming duckbill presents a third, ap-
THE CASE OF THE DUCKBILL'S HAND I 149
Duckbill ancestors had long toes.
Dryosaurus and Laosaurus (shown
here) had much longer hind toes
than did the more advanced
duckbills.
parently very strong point—the flattened tail. Crocodiles and Nile
monitor lizards are excellent swimmers, which employ the sculling
strokes of their deep, flat-sided tails to propel themselves through
tropical lakes and rivers. Deep, flat-sided tails are the characteris-
tic locomotor equipment of other reptilian swimmers, too—the sea
snakes of Indonesia and the extinct giant sea lizards of the Creta-
ceous. When the first complete duckbill skeletons turned up in the
1880s, one striking peculiarity was immediately noticed: the un-
precedented depth of the tail vertebrae. The bony spikes rising
150 I THE HABITAT OF THE DINOSAURS
from the upper vertebral bodies were of great height, and the lower
vertebral spines, called chevrons, were nearly as long. And so it
was proved: duckbills swam with sculling movements of their deep
caudal organ. And nearly all the textbooks repeat this conclusion
today, with complete assurance. But this hypothesis is rendered
highly dubious by every important detail of the anatomy.
First, there's the problem of muscle power to support the al-
leged swimming stroke. Swinging the flat-sided tail back and forth
against the water's resistance would require great muscular power.
Strong tail-scullers today have thick muscles at the base of the tail
and great spines of bone, called transverse processes, beneath them,
sticking out sideways from the vertebral bodies to provide strong
sites for the attachment of the muscles. Wide transverse vertebral
processes are found in some dinosaurs—the armored ankylosaurs
possessed outstandingly wide tail bases and must have had great
power in the sideways swing of their tails. As the ankylosaur's tail
ended in a bony war club, it's not surprising that the muscles at
the base of that tail were provided with strong attachments. The
duckbill's ancestors were also fairly strong in the rump; the Dry-
osaurus clan show good-sized transverse processes on the first ten
tail vertebrae, counting from the hips back. Since evolution sup-
posedly made duckbills better swimmers than their ancestors, we
should expect to find duckbills outfitted with very wide transverse
bony spines. We find no such thing. Edmontosaurus, the wide-
mouthed duckbill, had a tail base of only moderate width, nar-
rower relative to its body size than that of the ancestral dryosaurs.
It would have required a massively muscled rump to send Edmon-
tosaurus sculling through the Cretaceous bayous, and its modest
tail base was manifestly inadequate for that.
There are even more serious problems. The most specialized
duckbills were the hollow-crested group, Corythosaurus, Parasau-
rolophus, and their kin. They are inevitably portrayed as water-
lovers with prodigious natatory prowess. But their transverse tail
spines were absolutely puny—short, thin, weakly braced prongs of
bone, which could have supported only an atrophied set of tail-
flexing muscles. Thus hollow-crested duckbills, allegedly among the
strongest swimmers in all dinosaurdom, were actually weak-rumped,
puny-tailed creatures incapable of the powerful contractions re-
quired of a fast-swimming sculler.
THE CASE OF THE DUCKBILL'S HAND I 151
Further problems concerning the caudal configurations are
raised by the geometry of the upper tail spines. The duckbill's up-
per spines are indeed very long. The tails of living crocodilians also
feature long spines. But the geometrical arrangement is totally dif-
ferent. Crocodile tail spines rise almost straight up, from the seg-
ments of the backbone. So do those of the Nile monitor lizards.
Most land-living lizards—the iguanas, for example—also possess tall
tail spines, but in them the bony spikes slant strongly backward.
The difference between vertical and slanted spines is explained by
Why duckbills couldn't swim well. A good tail swimmer—like a modern
crocodile—has a special arrangement of bony prongs in the backbone. Wide
prongs that stick sideways attach to the powerful tail muscles. And tall
vertical prongs give sinuous flexibility. Duckbill dinosaur tails were wrong on
both counts. Their sideways prongs (transverse processes) were too short to
support
big muscles. And the upper prongs (neural spines) were too short
and slanted too far to the rear.
152 | THE HABITAT OF THE DINOSAURS
ligament action. A layer of tough ligament runs along the midline
of the tail, and each spine is embedded into it. When the spines
slant backward, this ligament is stretched when the tail is bent side-
ways. Such stretching impels the tail back in the direction from
which it came, so the land lizard can flex its tail back and forth
quickly as it runs. Underwater, much slower, stronger flexing would
be required, and the elastic-rebound effect produced by slanted
spines would be useless. So specialized tails of swimmers have
vertical spines.
Duckbills were supposedly croc-style swimmers, moving by
strong, easy, side-to-side flexures of their tail. Therefore, the op-
timal design would feature vertical tail spines. But duckbill spines
all slanted strongly backward, exactly as in land-living lizards, not
in swimmers.
Another problem in the duckbill's swimming equipment lies
in the profile of the tail. The deepest part of the croc's tail is close
to the end, because the end swings through a wider arc than does
the base in moving side to side. Thus the tail is deepest where it
can do the most good in pushing against the water. All powerful
tail-scullers have such deep tail ends. But duckbill tails were deep-
est at the hips and become progressively narrower from top-to-
bottom toward the tip—another caudal feature nearly totally
maladapted for its alleged primary function.
An argument very eloquently expressed in 1964 by John Os-
trom of Yale administers the coup de grace to the theory that
duckbills swam. Any sort of tail-propelled swimming requires a
smooth ripple of tail flexure from the hip out to the tip of the tail,
a sort of muscular sine wave that pushes against the water's resis-
tance to propel the animal forward. Even with the handicap of a
weak rump and their maladapted shape, duckbills could have swum
at least at slow speed if they could undulate their tails. The only
anatomical feature that would have entirely prevented a dinosaur
from swimming would be a tail corset, a stiff latticework of bone
which would hold the entire tail assembly together as one stiff im-
mobile mass. Tail corsets evolved in a wide range of dinosaurs,
including some meat-eaters, some plant-eaters, and the long-tailed
flying dinosaurs (pterosaurs). No modern reptile has a tail cor-
set—all the lizards, turtles, and crocs can wiggle their tails freely,
and even the weak-tailed lizards (desert horned toads, for exam-
THE CASE OF THE DUCKBILL'S HAND
153
pie) can employ their caudal undulations to swim when the animal
is forced to. But when the skeletons of duckbill tails were found,
their single most startling feature was a basketwork of long, stiff
bony rods crisscrossing over the backbone from the mid-chest all
the way to near the tail's end. A perfect tail corset. Each rod con-
sisted of dense bone up to half an inch thick and was attached at
one end to the bones of the vertebral column by a short stiff lig-
ament. Dryosaurs had a tail corset, too, as did horned dinosaurs,
but the ultimate in caudal basketwork was developed in the duck-
bills. *
The duckbill's tail corset evolved for an obvious mechanical
purpose: to keep its backbone stiff and immobile from a point just
behind its shoulders all the way down to the hindmost tail section.
Even the most devout believers in swimming duckbills are forced
to admit that this bony latticework would make for an unusually
unsupple spine, the very reverse of what is necessary for swim-
ming with smooth, horizontal undulations.
The supposedly definitive monograph on duckbills came out
in 1942. Its two authors (one was the senior professor of paleon-
tology at Yale) had to engage in quite a twisted form of logic to
explain away the problem of the duckbill's stiff backside. They ad-
mitted the bony system of rods must have evolved to maintain the
backbone rigid for perambulations on land when the duckbills chose
to walk about on terra firma. But perhaps, the authors argued, the
tendons were a little loose so that a small degree of side-to-side
movement was possible in the tail. But they ignored a critical
problem. The duckbill's ancestors had been land livers, with mod-
erately strong tail corsets, and the duckbills themselves increased
the stiffness of the corset. The monographers of '42 failed to ex-
plain why duckbills would evolve in the wrong direction—why the
duckbill family had stiffer, not more supple, tails than their ances-
tors.
The sum of evolutionary evidence is thoroughly damning. In
nearly every modification of the evolutionary process made in the
duckbills as they developed from their dryosaur ancestors, the
duckbills suffered a diminution of their swimming potential. Their
fore- and hind paws became shorter and more compact, not longer
and more widely spread. Their tails got weaker and stiffer. Far from
being the best, the duckbills must have been the clumsiest and
154 | THE HABITAT OF THE DINOSAURS
Body-tail corset
of a duckbill
(Corytbosaurus)
slowest swimmers in all the Dinosauria. If pressed, they probably
could paddle slowly from one riverbank to another. The central
theme of their bodily evolution was indeed specialized—orthodox
theory was right on that point—but the direction of specialization
was landward. These dinosaurs were specialized for a totally ter-
restrial existence.
Every so often some paleontologists attempt placing some other
major dinosaur group in the water. A young Canadian would have
Alberta horned dinosaurs wading through the sluggish backwaters
of the Judith Delta. But his evidence derived from the dubious
notion that the horned dinosaurs spent most of their lives in the
water because their fossils are found buried in river-channel sand-
stone. American buffalo often are found buried in river sands where
their bodies came to rest after a flood, yet the buffalo is hardly an
THE CASE OF THE DUCKBILL'S HAND
155
aquatic creature. One quick way to calculate how well a dinosaur
would cope with swampy terrain is to calculate the area of its hind
foot available to support the downward thrust of its hind leg. Any-
one who has tramped around as many bogs and swamps as I have
knows that feet get stuck not when you stand still, but when you
step too forcefully and drive your leg down into the muck. The
faster you walk, the more downward thrust is applied to the sole
of the foot. Hence to move speedily over mud or soft earth, de-
vices such as snowshoes must be used to expand the foot's area.
Hippos follow this pattern: they have wide-spreading toes relative
to the power of their legs, a contrast to the small, short-toed feet
of the rhino. Since all dinosaurs had stronger hind limbs than fore,
the
largest thrust would be exerted by their thigh muscles. So the
thickness of the thigh bone (femur) works as a useful gauge of the
Long toes in a hippo (left) and
short toes in a brontosaur (right)
156
THE HABITAT OF THE DINOSAURS
force applied to the sole of the feet. From fossil footprints, the
area of a dinosaur's foot can be calculated and then compared with
the cross section of the femur.
This exercise in quantitative paleopodiatry produces dis-
tinctly counterorthodox conclusions. All the popular books list
- brontosaurs and duckbills at the top of the list in the preference
for swamps. But brontosaurs and duckbills had among the small-
est feet in area relative to the size of their thigh. If these giants
had tried to spend their lives paddling around marshy terrain, they'd
have found themselves stuck in the mud with genuinely maladap-
tive frequency.
Stegosaurs and ankylosaurs were also compact of foot. But the
horned dinosaurs had much bigger feet per pound of thrust from
the thigh. A few dinosaurs were especially large-footed—some lit-
tle horned dinosaurs, the primitive anchisaurs (brontosaur ances-
tors), the dryosaurs, and most of the meat-eaters, both large and
small. Strangely enough, it's these dryosaurs and meat-eaters that
are supposedly least adapted for soft swampy soils according to
orthodox dinosaur ecology. But like so much else in traditional
dinolore, the standard story about feet and mud is not accurate.
Museum exhibits teach that brontosaurs and duckbills escaped their
predatory enemies by wading out into the marshes where meat-
eaters feared to tread. But if an allosaur pursued a brontosaur or
a tyrannosaur pursued a duckbill, it would be the ponderous pads
of the plant-eaters that would mire first into the mud to hold their
hapless owners fast as the killer descended.
After all this calculation of tail mechanics and foot areas has
been done, that duckbill mummy's hand, that webbed forefoot
raised forever skyward in its fourth-floor glass case at the Ameri-
can Museum of Natural History still requires an explanation. If it
was not for swimming, then what was it for? A dead camel I ob-
served in the Transvaal might solve this mystery. While I was in
Robert T Bakker Page 16