right. Protecting oneself from a predator was always a vital func-
tion, and any anatomical device that could be wielded as a ram or
club was useful in that connection. But in the evolutionary scheme
of things, staying alive means nothing unless one's genes are passed
on to the next generation. So, the ability to butt a sexual rival hard
in the ribs might provide an edge in the great evolutionary dating
game.
Without doubt the most dangerous devices for active defense
among the Dinosauria emerged in Triceratops. The scene has been
portrayed in paintings, drawings, and illustrations hundreds of times,
but it remains thrilling. Tyrannosaurus, the greatest dinosaur to-
reador, confronts Triceratops, the greatest set of dinosaur horns.
No matchup between predator and prey has ever been more dra-
matic. It's somehow fitting that those two massive antagonists lived
out their co-evolutionary belligerence through the very last days
of the very last epoch in the Age of Dinosaurs. Tyrannosaurus stood
240 I DEFENSE, LOCOMOTION, AND THE CASE FOR WARM-BLOODED DINOSAURS
over twenty feet tall when fully erect, and a large adult was as heavy
as a small elephant—five tons. No predatory dinosaur, no preda-
tory land animal of any sort, had more powerful jaws. Withstand-
ing a Tyrannosaurus's attack required either tanklike armor—the
approach taken by Ankylosaurus—or most powerful defensive
weapons—the approach taken by Triceratops.
Triceratops's body was designed for lunging and charging. The
torso was very short, the chest broad, the hips wide and strong.
Fore- and hind limbs were very thick for the body size—much
thicker than an elephant's of the same weight—and the paws were
wide and compact. No armor plate encased the hide of Tricera-
tops, because its defense was active, head-first, and devastatingly
effective. Triceratops and its kin carried far and away the largest
and heaviest skull ever to evolve on a land creature—six, seven,
even eight feet long, up to four feet wide, and of very solid con-
struction. Where the neck muscles attached to the back of the skull,
the cranial bones had expanded sideways and upward and were
reinforced to support sudden twisting lunges of the great horns
located on the brows.
Triceratops's horns were wonderful examples of Mesozoic ar-
mature. From the eye socket to tip, the horn cores could reach
four feet in length and often had a graceful double curve like the
horns of longhorn cattle from the Wild West. When the first Tri-
ceratops horns were discovered in Colorado in 1880, Professor
Marsh thought they had belonged to ancient buffalo. But although
Triceratops horns were shaped like a buffalo's, they were located
on the head in a far more dangerous orientation. Longhorn cattle
and buffalo horns face sideways, and their horn thrusts can only
be delivered by tossing the. head to the left or right. Triceratops
thrusts could be far more precise. Its horns curved forward and
slightly outward over the long snout. Although the head was mas-
sive, it was nearly perfectly balanced on the ball-and-socket-type
joint between the head and the neck. The heavy snout forward
was counterbalanced by the broad head shield extending back-
ward. The entire apparatus was a marvelous combination of deli-
cate musculoskeletal poise and brute power, allowing Triceratops
to lunge forward at its opponent with the entire set of horn tips.
The neck is a vulnerable point in any vertebrate, and Tricer-
atops protected its neck with a flaring collar of bone, fringed by
MESOZOIC ARMS RACE I 241
Two Triceratops confront a Tyrannosaurus.
Triceratops drumsticks. Muscular power for quick charges came from the
huge calf muscles that attached to the inner and outer sides of the big bony
crest of the shin (cnemial crest). This same crest also gave the knee-
straightening muscles great leverage.
short, sharp, horn-covered spikes. Part of this frill was covered by
an extension of the jaw muscles. But the wide periphery of the
bony frill was pure armor, covered with tough horny skin. Below
each eye and just above the jaw joint was a short horn-covered
spike that protected the cheek. This defensive master-machine alive
and in action must have been a sight to behold, its eight-foot skull
pivoting easily left and right, its neck frill swinging in wide arcs.
Triceratops was not the only giant horned dinosaur found on
the Laramie Deltas. It was accompanied by the rare Torosaurus—
the "bull lizard"—which sported an even longer neck frill. And in
New Mexico, during Late Cretaceous days, a splendid long-horned
relative of Torosaurus walked the floodplain— Pentaceratops, the "five-
horned face," named for its combination of unusually long cheek
horns, brow horns, and nasal horn.
The finest display of horned dinosaur heads anywhere in the
world is located in the Cretaceous Hall of the American Museum
of Natural History in New York. The horned legions reached their
greatest variety during Judithan times, a few million years earlier
than the age of Triceratops. Judith River beds in Montana and Al-
berta have been very generous to dinosaurophiles. The New York
244 | DEFENSE, LOCOMOTION, AND THE CASE FOR WARM-BLOODED DINOSAURS
Perfect balance at the
Triceratops skull pivot.
Horned dinosaurs had a
ball-and-socket joint
connecting the massive head
to the neck. Since the joint
was placed under the eyes,
just at the natural balancing
point of the huge head, the
neck muscles could toss the
head in any direction with
great precision of
movement. Botton cutaway
view shows the ball joint on
the skull separated from the
socket.
museum displays Centrosaurus, a. short-frilled variety whose weak
brow horns were compensated by an erect and very long horn over
the snout. Its close kin Styracosaurus possessed the great nose horn
plus a magnificent set of curved spikes over the frill, giving its head
a monumentally prickly appearance. Monoclonius is there too, with
its stout nose horn. In general, Judith River horned dinosaurs sort
out into two systems of attack. The first includes the genera with
huge nose horns and weaker brow horns. These animals probably
thrust their powerfully armed snouts straight upward as they tried
to gore the softer underparts of a tyrannosaur. The second system
includes the Triceratops-like configurations—long brow horns
curving forward. Such long brow horns are rare among the Judith
fauna—but the species Chasmosaurus kaiseni is there to represent
this second system.
MESOZOIC ARMS RACE 245
In the 1890s, horned dinosaurs confronted science with an
evolutionary puzzle: These dinosaurs were so highly evolved for
an aggressive defense that paleontologists were at a loss as to how
such creatures could have descended from any other kin
d of beaked
Dinosauria. Even the oldest horned dinosaur fossils from North
America manifested the very complexly designed snout, horns, frill,
and neck muscle attachments in a fully developed state. It was as
though the horned dinosaurs had sprung directly from the mind
of the Creator.
Today the early fossil record of the horned dinosaurs is still
imperfect, but we are two big steps closer to understanding their
evolutionary origin, owing to discoveries in the Cretaceous sands
of the Gobi Desert. The spectacular discoveries made in the Gobi
came about through a colossal error of scientific theory. American
scientists in the early 1900s wanted to explore the Mongolian des-
ert because a theory popular at the time maintained that human
evolution occurred fastest and most efficiently on a dry, invigorat-
ing plateau such as the country of Central Asia—the "Roof of the
World." Conversely, tropical lowlands were supposedly evolution-
ary slums where stagnant water and fetid air suppressed the de-
velopment of higher life forms. According to this theory, only where
air was dry and thin—as on the Asian Plateau—could lively spe-
cies evolve. These Asian plateaus were terra incognita at the time;
no thorough scientific surveys of Gobi zoology or paleontology had
been made, and no one knew what sort of beasts had evolved there.
After the first World War, wealthy Americans contributed funds
for a grand American Museum expedition to the Gobi. The trip
was billed as the search for the missing link; the key to human
evolution was to be found in the windswept desert. Roy Chapman
Andrews, naturalist-explorer par excellence, was its leader.
So far as missing links in evolution were concerned, the ex-
pedition was a bust. No important protohuman fossils were found.
We now know in fact that nearly all the steps in human evolution
took place in warm tropical realms, not on high plateaus. But the
Gobi expedition uncovered a boundless treasure trove of dino-
saurs, whole new families of them. These Gobi Cretaceous dino-
saur beds were totally different from the Judith and Laramie Deltas
familiar to the American geologists. As we have already noted, Late
Cretaceous habitats in America were mostly humid deltas, but
248 | DEFENSE, LOCOMOTION, AND THE CASE FOR WARM-BLOODED DINOSAURS
Central Asian habitats of the time were dominated by desert and
near-desert conditions. Red Gobi sandstones preserved the sedi-
mentary work of Cretaceous winds that drove sand into dunes
around shallow lakes. Over millions of years these Cretaceous dunes
coalesced into dune fields, and these fields, in turn, piled upon one
another to produce hundred-mile-wide layers of preserving sand.
Andrews's field parties found innumerable white skeletons in the
red rock—small, chunky-bodied dinosaurs with long hind legs,
powerful beaks, and short frills protecting their necks. Andrews's
men had found primitive horned dinosaurs, the ancient uncles of
Centrosaurus and Triceratops.
When the scientists in New York unpacked the first crate-
loads of dune rock from the Gobi, it was clear a dinosaur missing
link had been found. The new dinosaur's name was a tribute to
Andrews's leadership: Protoceratops andrewsi, "Andrews's ancestral
horned-face." Protoceratops's cranial structure was almost perfect as
the ancestral state of the large American horned dinosaurs. The
basic horned dinosaur design was proclaimed by the deep beak,
solidly connected skull bones, and a well-braced neck frill. But
Protoceratops displayed only the suggestion of horns. In the biggest
skulls a roughened bump on the snout must have supported a low
horny crest in life. And Protoceratops1 % legs, hips, and shoulders were
delicate compared to the massive strength of the American horned
dinosaurs.
Protoceratops and its close relatives must have swarmed over
Asia, because their bones and nests of eggs are the commonest
dinosaur fossils found in the widespread dune beds. But not one
Protoceratops has ever been reported from the rich beds of the
American Judith and Laramie Deltas. Swampy meadows and broad
humid floodplains were evidently not to Protoceratops1 % liking, though
Canada and Montana did play host to relatives in Late Cretaceous
times—the genera Leptoceratops ("diminutive horn-face") and Mon-
tanoceratops. Leptoceratops probably was an immigrant from Asia.
Roy Chapman Andrews's team also discovered a second miss-
ing link in horned-dinosaur history: an earlier Mongolian family
which at last revealed how the horned-dinosaur story began. At
first the announcement created little stir—two skeletons and some
odd bones from a very small beaked dinosaur out of the Early
Cretaceous beds of Inner Mongolia. (One of the skeletons had
MESOZOIC ARMS RACE I 249
Protoceratops: male (left) and female (right)
nicely preserved gizzard stones.) But before long dinosaur anato-
mists began to discover what they had; these tiny beaked dino-
saurs possessed a very deep, parrotlike snout that looked just like
that of Protoceratops. It was this beak that suggested the name for
the beak, but the rest of this parrot dinosaur's skull began to look
suspiciously like that of an ancestor for Protoceratops. Parrot dino-
saurs protected their cheeks with sharp, crestlike horns, and they
showed just the beginnings of a neck frill. Especially striking to
the anatomist's eye was the core of the upper beak; it was formed
by a bone separate from the rest of the skull, a most unusual trait,
found elsewhere only in true horned dinosaurs.
Thanks to these Asian discoveries, and those made in recent
decades by Polish, Russian, and Chinese expeditions, we now pos-
sess an outline of horned-dinosaur history. Parrot dinosaurs must
have been close to the original ancestral stock—they have the ex-
tra-long hind legs and short forelimbs so common among primi-
tive beaked dinosaurs of all sorts. In parrot dinosaurs we already
find the trend toward an exceptionally strong head with a power-
ful beak and strong bite. Parrot dinosaurs were leaf-eaters, it's clear
from their teeth. But their beak served both as an herbivory and
an antipredator device, equally snipping off branches and snap-
ping menacingly at predators that threatened attack. The crest-ar-
mored cheek bones would protect the parrot dinosaurs when they
lunged to bite at an enemy. From such a beginning it was only a
short evolutionary step to Protoceratops and Leptoceratops with their
incipient horns.
The horned-dinosaur story shows how paleontologists can trace
the major evolutionary events. Rarely do fossils yield a complete
evolutionary sequence from mother to daughter to granddaughter
species. Evolution is too bushy to permit such a straightforward
story, too full of side branches. As clans evolved, the ever-branch-
ing species spread over the continents. Since fossils
come from a
few small areas, it is impossible to follow every stage of an evo-
lutionary line. But it's possible to make out an overall progression
of uncles and nieces, even when the parent—daughter sequences
cannot be found. Parrot dinosaurs probably weren't the direct
ancestors of Protoceratops, since the parrot clan had already branched
off in their own unique direction (parrot dinosaurs had evolved
MESOZOIC ARMS RACE | 251
Three Psittacosaurus,
parrot dinosaurs,
flee a predator.
very narrow forepaws with fewer fingers than true horned dino-
saurs). And Protoceratops probably wasn't the direct ancestor of
Triceratops or Monoclonius. Protoceratops and its sibling genus Lep-
toceratops had evolved into an evolutionary sideline where the tail
had become very slender from side to side but quite tall from top
to bottom. This tall-tailed condition was probably part of Protocer-
atops's tactics of intimidation—broadside huff and bluff. The big
North American horned dinosaurs had neither the narrow paws of
parrot dinosaurs not the billboard-type tails of Protoceratops. To sum
up, parrot dinosaurs were the granduncles of Protoceratops, which
was the granduncle of Triceratops.
Altogether the history of dinosaur arms and armor must rank
as one of the most dramatic aspects in the pageant of evolution,
but it poses a question. Was this parade of tanks and dread-
noughts one natural unit of evolution, a single major branch of
252 | DEFENSE, LOCOMOTION, AND THE CASE FOR WARM-BLOODED DINOSAURS
the dinosaur family tree, or not? In the early days of American
dinosaurology, museum scientists believed that stegosaurs, nodo-
saurs, and ankylosaurs were closely related and that horned dino-
saurs were perhaps cousins. But this idea lost favor among
Americans in the 1920s for no good reason. And for the last half
century most books place the armored groups—stegosaurs, anky-
losaurs, and horned dinosaurs—into separate suborders.
On the other side of the Atlantic, a very different conception
was adumbrated by a flamboyant but perceptive Hungarian pa-
leontologist, the Baron von Nopsca. Nopsca was a bright spot