A Cold, Hard Look at Dinosaurs
By Virginia Morell

A decade ago the dinosaurs got a makeover and were miraculously transformed from cold-blooded
sluggards to warm-blooded bon vivants. Now a group of researchers wants to cool then down
again--but still keep them interesting.

Golden eye, a two-foot-long savanna monitor lizard, is about to make a contribution to science. His
keeper, Tomasz Owerkowicz, sets him down on a treadmill enclosed by a large plastic box, then flips
the power switch. Golden Eye's pink, forked tongue flicks out, testing the surface of the treadmill.
When Owerkowicz nudges him with his gloved hand, Golden Eye starts to trot at a steady pace. "He's
my best runner," says Owerkowicz proudly. Occasionally he offers the lizard
encouragement--sometimes another nudge, sometimes a cheer ("Come on, sweetie pie"). Soon
Golden Eye is trotting faster and his throat is moving up and down. "I used to think that was
connected with the tongue-flicking, but it's not," Owerkowicz says. "They do it when they get tired;
it's a way to increase their oxygen intake, I think."

Owerkowicz, a graduate student in biology at Harvard, is happy to see Golden Eye getting a good
workout. He is something of an animal personal trainer--every day for the last few months, he has
coached savanna monitors, hedgehogs, and ground squirrels through 20-minute treadmill workouts. In
their well-exercised bodies (and particularly in their bones), Owerkowicz thinks there are clues to one
of paleontology's longest, hottest, and most popular debates: Were dinosaurs warm-blooded
endotherms like the hedgehogs and squirrels, or cold-blooded ectotherms like Golden Eye?

In the last decade, warm-bloodedness has gained a firm upper hand. Remember the scene in Jurassic
Park in which high-energy velociraptors let out steamy breaths as they stepped into a walk-in
refrigerator? Cold-blooded animals could never have done that. And though clearly the scene is
Hollywood make-believe, many scientists think it has a core of truth. Over the years they've amassed a
pile of evidence-- everything from the microscopic structure of fossilized dinosaur bones to the
dinosaurs' postures--to support the case for warm-bloodedness, and their argument has been
persuasive. Textbooks now usually portray dinosaurs as hot to trot, and natural history museums are
furiously revamping their displays to reflect the image makeover. Once thought of as dim-witted,
sluggish, oversize reptiles, dinosaurs are now shown as highly intelligent speedsters. Triceratops eludes
its predators by galloping away; herds of migrating duck-billed dinosaurs called hadrosaurs care for
their young in communal nesting grounds; and a number of carnivores, including Velociraptor, leap
and slash at prey with lethal, sickle-shaped claws.

Yet not everyone has been lured into the warm-blooded camp. Even as the museum technicians were
starting to hike up their fossils, a small but growing coterie of paleontologists and physiologists was
preparing to launch an assault against the new vision of the extinct beasts. They began questioning the
claims being drawn from the fossil evidence, searching for a truly incontestable mark of endothermy.
So far, these revisionists have found, all the studies suggest that dinosaurs should go back to the
ectothermic fold.

"A lot of this business about warm-blooded dinosaurs was just blown way out of proportion," says
Alan Feduccia, an evolutionary biologist at the University of North Carolina at Chapel Hill. "It was
just pure hype. So I applaud these new studies. Dinosaurs are beginning again to look like
reptiles--although not your everyday ones." As Owerkowicz and crew are quick to point out, a reptilian
metabolism is not an insult. "Dinosaurs could still be very active and agile, especially the bipedal
ones," says Owerkowicz, "and they could have lots of interesting behaviors, just as in Jurassic Park.
But no, I don't think they would have steamed up the refrigerator."

The debate over dinosaur metabolism is almost as old as the study of dinosaurs itself. In 1825 the
public was stunned by reports that newly discovered teeth and bone fragments had come from a reptile
twice the size of an elephant. By 1841, after more dinosaur fossils had been found, the British
anatomist Richard Owen decided that the extinct beasts were distinctive enough to warrant their own
suborder, which he called Dinosauria (for "terrible lizards"). He considered them "the crowns of
reptilian creation," so vivacious and energetic that they were "the nearest approach to mammals."

Despite Owen's recognition of the dinosaurs' unusual nature, for more than a century they were
lumped together with other reptiles and inevitably depicted as low-slung, oversize lizards with sprawled
legs and dragging tails. Besides, the idea of a giant cold-blooded reptile made perfect sense: in the Age
of Dinosaurs, from about 228 to 65 million years ago, the Earth was thought to be a steamy hothouse.
Swampland bordered warm seas, and conifers and tree ferns shot skyward. Since the air would keep
dinosaurs warm, there was no need for them to be charged-up endotherms.

Only in the late 1960s was this idea seriously challenged. John Ostrom, a paleontologist at Yale, and
Robert Bakker, then a Yale undergraduate, concluded independently that the standard depiction of
dinosaurs was wrong. Applying the laws of biomechanics to fossilized bones, Ostrom and Bakker
argued that contrary to the then prevailing imagery, dinosaurs had their legs squarely under their
shoulders and hips and held their tails stiffly in the air. Ostrom drew particular inspiration from
Deinonychus, a Velociraptor-like dinosaur he found in 1964, whose foot ended in a long, thin,
scythe-like claw. Here was an animal built for speed and agility--of a sort that today's reptiles seem not
to possess--that practically demanded a warm-blooded metabolism to power it. "It must have been a
fleet-footed, highly predaceous, extremely agile, and very active animal," he wrote. "These in turn
indicate an unusual level of activity for a reptile and suggest an unusually high metabolic rate." In
other words, Ostrom doubted that Deinonychus was a standard ectotherm, depending on the warmth of
the sun for its energy.

This is exactly the strategy that almost all living fish, amphibians, and reptiles use. Lacking a
physiology that can raise or lower their body temperatures much above or below that of the
surrounding air or water, they fluctuate with their environment. Ectotherms are not passive and helpless,
though; to warm up, they may bask in the sun; to cool down, they can retreat to the shade.
Warm-blooded endotherms, on the other hand, maintain a relatively constant internal body temperature
night and day, winter and summer. In mammals, it hovers between roughly 97 and 99 degrees
Fahrenheit; in birds, about 104. Their higher body temperatures (relative to those of ectotherms) mean
that the chemical reactions that make up an animal's metabolism can work faster, which makes a more
active way of life possible. On average, endotherms have a metabolic rate four times higher than
ectotherms of the same size. After a cool night, a savanna monitor lizard has to wait for the morning
sun before springing into action. A lion doesn't. Nor, Ostrom hinted, did Deinonychus.

Bakker was more explicit, arguing for a complete overhaul of our understanding of dinosaur
physiology and behavior. Endothermy, he said, gave dinosaurs the speed, energy, and other gifts that
allowed them to dominate the land for more than 100 million years. By the mid-1980s he was no
longer alone. A number of other paleontologists were impressed by evidence such as the dinosaur
nesting sites complete with eggs, embryos, and young found by Jack Horner of Montana State
University. Horner's initial analysis of these bones suggested that the baby dinos were growing as
rapidly as young ostriches. And that kind of speedy growth, like the agility of Deinonychus, seemed to
be a hallmark of endothermy. At the same time, other researchers were constucting a tree of life in
which birds evolved from dinosaurs much like Deinonychus. Since birds are warm-blooded, it seemed
reasonable to suggest that their immediate ancestors might have been as well.

All these arguments for dinosaur endothermy suffered, however, from the same defect: they were
indirect. As a result, dinosaur researchers began making pilgrimages to Paris to visit Armand de
Ricqles, a paleontologist and anatomist at the University of Paris who claimed that he could tell the
difference between endothermic and ectothermic animals by putting their bones under a microscope.
Here, at last, seemed to be a concrete method for resolving the debate. When De Ricqles looked at
tissue- thin slices of dinosaur bone, he found striking similarities to the bone of birds and mammals. In
young warm-blooded animals today, fast-growing bone invades connective tissue so quickly that it
traps the fibers and blood vessels already there in a dense, intricate weave. As that bone matures,
channels known as haversian canals appear, in which specialized cells destroy old bone and replace it
with new material. De Ricqles argued that the texture of the bone and the presence of the canals could
be used to deduce that an animal's skeleton was growing quickly. And since fast growth naturally
required a high metabolic rate to fuel it, such an animal had to be endothermic.

A reptile's skeleton, in contrast, is made of more orderly, layered bone with relatively few haversian
canals. Bones of cold-blooded animals also often contain growth rings much like the annual growth
rings of trees. As in trees, they mark the periods when the animals slowed their growth or stopped it
altogether. De Ricqles argued that the dinosaur bones he examined had the same pattern of canals as
birds and mammals, with none of the growth rings of reptiles. Thus, he concluded, dinosaurs were
probably warm-blooded.

"De Ricqles's theory became very entrenched in the paleontological world," says Owerkowicz,
"because it tells you exactly: to judge extinct animals, all you have to do is look at the bone." Actually,
by the 1980s De Ricqles had modified his theory. He noted that some dinosaur bones do contain
reptile-like growth rings and that it was possible they weren't exactly endothermic. Still, he remained
convinced that bone could be used as an indicator of an animal's thermoregulation.

Not all paleontologists bought the equation. "Fuzz" Crompton at Harvard's Museum of Comparative
Zoology was one of the doubters. And when Owerkowicz showed up in his lab five years ago,
Crompton suggested that the young man put De Ricqles's theory to the test.

"The funny thing is that when I started my research, I believed De Ricqles," Owerkowicz says,
shaking his head and shrugging. But when he first arrived in the United States, the Polish-born
physiologist did not question authority. "I come from a very strong Catholic family and a Communist
country, where you were taught to believe what you read," he says. "I only learned to question things
after I'd been here. And in a way that was good. It shows I wasn't hell-bent on disproving De
Ricqles's hypothesis." At first, it even shocked him that Crompton raised some doubts about De
Ricqles's finely documented work. But now, with his mind "liberated," as he puts it, Owerkowicz is
quite comfortable pointing out weaknesses in his elders' research and making blunt pronouncements
of his own.

"De Ricqles's theory did a lot of good," says Owerkowicz, "because it showed people that you could
use the tissues of the fossils to mean something, which no one had done before. But I think he took it a
little too far." What De Ricqles had not done, and what Crompton encouraged Owerkowicz to do, was
test the idea experimentally in living animals--both endotherms and ectotherms.

In 1993, Owerkowicz began acquiring savanna monitor lizards, including Golden Eye, as well as
hedgehogs and ground squirrels. He chose the animals in part because they are about the same size. (A
mouse and an elephant are both warm-blooded, yet the mouse, with a much larger surface area
compared with its mass, has a much higher metabolic rate.) He also chose his animals because they
were at three distinct points on the warm- to-cold-blooded spectrum. The ectothermic lizards were at
one end, the endothermic ground squirrels at the other, and the hedgehogs were close to the middle:
although they are endotherms, they consume oxygen at a very low rate for a warm-blooded animal.
Owerkowicz divided each species into a sedentary, "couch potato" set and a workout crew. All lived
with the same light and heat, which was enough for the lizards to maintain a constant body temperature.

For six months, he exercised each animal in the workout group for 20 minutes every day, blasting away
the boredom with loud Polish rock music (Budka Suflera is a favorite group). Every six weeks he gave
couch potatoes and workout crew alike an injection of fluorescent dye. As the animals subsequently
formed new bone, some of the dye was deposited within the crystals of the bone. With each injection
of dye, Owerkowicz was creating a timescale for the growth of the animals' bones, and he could use it
to calculate how fast the bones were growing. That's something that De Ricqles could only infer from,
not observe in, his own work. At the same time, Owerkowicz could tell which regions of bone tissue
were the result of new growth and which the result of simple remodeling.

To see this scale, unfortunately, it was necessary, as Owerkowicz puts it, "to bump the animals off.
And it's hard, because you grow attached to them." Afterward, he sealed the bones in plastic, placed
each one in a vise, and with a diamond-blade saw sliced it into one-millimeter-thick cross sections,
which he fixed on slides.

After months of nudging animals on treadmills and slicing up their bones, Owerkowicz now has a
stack of neatly labeled slide boxes, whose contents hold a challenge for De Ricqles and the
warm-blooded dinosaur school that relies on his methods. Opening one, Owerkowicz selects a slide,
secures it under a microscope's prongs, and adjusts the focus. "This is a femur of a hedgehog," he
says. "I gave it four dye injections-- greenish yellow, red, orange, and then another greenish yellow."
Under the microscope, the thin section looks like an oval of spun crystal. One can't help exclaiming in
awe, since the colored bone is as delicate and lacily beautiful as a dragonfly's iridescent wing.

Owerkowicz discovered that the differences between the bones of different animals were nowhere near
as dramatic as one might assume. "When I compared the femurs and humeri of my exercised species,
I found that the deposition rate of bone was similar," he says. Nor was there a difference in the growth
rates of the bones of the sedentary animals. "When I kept my lizards at 95 degrees, they grew at the
same rate as the hedgehogs, even though their resting metabolic rate is five times lower, which tells me
their enzymes are capable of working just as fast," he says. "But they don't normally because they
don't need to; they don't have the high energy requirements to keep their internal body temperature up,
the way endotherms do."

The only clear distinction he could draw, in fact, was between the exercised and unexercised animals.
The ones that ran on a treadmill had bones packed with haversian canals--regardless of whether they
were warm- blooded or cold. "What that means," says Owerkowicz, "is that an animal's thermal
physiology doesn't shape a bone's microstructure. But exercise does."

When an animal exercises, he points out, its bones are strained and stressed. Since they are made of an
elastic material, they can generally withstand these forces, but occasionally they yield to fatigue and
form a tiny stress fracture. "That relieves the strain, but if the loading continues, it can lead to a full
fracture," says Owerkowicz. "The bone needs to remove that microcrack and put new bone in its
place. This is the role of remodeling. Bone that is dense with haversian canals, therefore, tells you that
the animal was active." What it doesn't tell you is the animal's resting metabolic rate or its body
temperature--in other words, whether it was warm- or cold-blooded.

The close study of dinosaur bones, Owerkowicz concludes, can never be the smoking gun
paleontologists once hoped it would be. "It might tell them about how dinosaurs grew, and how their
bone tissues changed while they were growing, and whether they were active or not. But I don't think it
will answer the big question. It won't tell them if the dinosaurs were endotherms."

While bone tissue may not answer the riddle of the dinosaurs, the shapes of the bones might. For John
Ruben, a physiologist at Oregon State University, the answer lies chiefly in their noses. "I didn't get
into this debate because I've got some ax to grind about dinosaurs," he says. "I got into it because I
built up a database on living animals and realized that there are some anatomic features associated with
endothermy that ectotherms don't have." These were features, he claims, that dinosaur paleontologists
have missed because they often know little about animal physiology-- particularly the physiology of
reptiles. "I'm constantly amazed that people haven't looked at some of these things," he says, barely
suppressing a gleeful smile. "I tell you, doing this research is like taking candy from a baby."

Ruben and a former graduate student, Willem Hillenius, who now teaches at the College of Charleston
in South Carolina, found that the noses of living animals could tell you a lot about their metabolism.
Ninety-nine percent of warm-blooded animals have coils of membrane-covered cartilage or bone in
their nasal passages called respiratory turbinates. The function of these structures was first discovered
in 1961 in desert kangaroo rats, and for a long time researchers thought they were useful only to
mammals living in arid conditions. When such animals breathe out moist, warm air from their lungs,
much of the water condenses onto the cooler turbinates. The animals then breathe in the dry desert air,
which picks up the water on the turbinates and brings it back down to the lungs. "Essentially, the water
is recycled back into the animal's respiratory tract without very much of a loss," explains Hillenius.

Yet almost all mammals and birds have turbinates, not just the ones that live in deserts. Hillenius and
Ruben therefore think that turbinates are useful for endothermy in any habitat. Mammals and birds
need turbinates, the researchers say, because they consume oxygen at a rate nearly 20 times that of
similar-size reptiles in order to fan the fires of their internal heaters. "Those high metabolic rates are
expensive," says Ruben. "They cost a lot in food and oxygen to maintain." The high rate of oxygen
consumption calls for a high rate of breathing--and that entails a high risk of losing water. Hillenius
and Ruben have found that without turbinates, a mammal loses 75 percent of its daily water intake.

In Ruben's lab, a group of jars filled with preserved, disembodied heads serves to illustrate their point.
A cow and a sea otter are on hand, their skulls cut away to reveal a profusion of turbinates, the otter's
so elaborate that they look almost like wavy coral. By contrast, a crocodile's skull reveals an empty
nasal cavity--not because any turbinates have dried up after death and fallen out but because crocodiles
simply don't need them and therefore don't have them.

Hillenius points to the slightly tattered and yellowed head of an ostrich that has been neatly sliced in
half, right down the midline. There, filling this warm-blooded bird's nasal cavity, is a mass of coiled,
cartilaginous turbinates. Turbinates are such a necessity for warm-blooded living, it appears, that the
structures have evolved in the two independent lineages of birds and mammals. Even though each
group uses a different set of cells in their embryonic stages to grow these humidifiers, the structures
end up in the same part of their skulls, looking nearly identical. "That says to me that it's difficult to
become an endotherm without some way of stopping this water loss," says Hillenius.

Having established what appears to be a positive correlation between turbinates and endothermy,
Hillenius next turned to the fossil record to see if he could find any evidence of these structures in
extinct animals. Because the turbinates in even recently dead animals are extremely fragile (in the
cow's skull, for example, they were tissue-paper thin), Hillenius doubted that they would be preserved.
So he looked instead for bony, slightly raised parallel ridges where the turbinate tissue attaches. And he
found them. Primitive insectivorous mammals 160 million years ago had large ones. Cynodonts,
mammalian ancestors some 250 million years old, had smaller ones. After that, the trail grew cold.
When he turned his turbinate-spotting eye on pelycosaurs, reptilian animals dating to 300 million years
ago, he found nary a ridge.

When Hillenius looked at birds, he found evidence of turbinates only as far back as 70 million years--a
little under half the time since they originated from dinosaurs. And when he finally looked at dinosaurs
themselves, he found none--which meant, he concluded, that the dinosaurs were not warm-blooded.

"I think these ridges are the first preserved feature that can be causally linked to metabolic rates and so
to endothermy," Hillenius says-- and a number of paleontologists agree. On the other hand, others
have remained skeptical that turbinates can be ruled out so categorically from some fossils.

To answer those critics, Hillenius, Ruben, and two of Ruben's graduate students have found a second
indicator of metabolism: the size of the nasal cavity itself. Teaming up with Andrew Leitch, an expert
on ct- scanning fossils, they measured the cross-sectional area of the cavities. In doing so, they've
found yet another strong distinction between endotherms and ectotherms. "We've done everything
from raccoons and coatimundis to black bears, humans, crocodiles, ostriches, and great blue herons,"
says Nicholas Geist, one of Ruben's students, placing the scan of an ostrich's skull on a light table.
"Here, right where the turbinates are present, there's an enormous amount of space, relative to the
bird's head size and its overall body size, that's devoted to the respiratory passage."

It's so big for two reasons: it needs to house large turbinates, and it has to handle the great volume of
air that the endothermic creature needs to inhale. A crocodile's skull, however, has a narrow nasal
passage, presumably because it needs less oxygen and doesn't have to fit in any turbinates. Overall, the
Oregon team found, the cross sections of nasal cavities of warm-blooded birds and mammals are four
times bigger than those of a similar-size reptile.

The researchers then started putting dinosaur skulls in their ct scanner. It was a tricky project, since few
dinosaur skulls are in good enough shape to be scanned. "They can't be squashed or distorted for this
technique to work," Geist says, "since we're tracing the interior region of the whole nasal cavity." So
far the team has scanned skulls of three dinosaurs: Nanotyrannus, a small relative of Tyrannosaurus
rex; an ostrichlike dinosaur called Ornithomimus; and a duck-billed dinosaur named Hypacrosaurus.
When they charted the ratios of the cross-sectional areas of nasal cavities to body mass for all the
animals they studied, the mammals and birds lined up in one neat row--and the dinosaurs fell right into
place with lizards and crocodiles.

Does all this research threaten to drag us back to a Victorian vision of dinosaurs? Ruben responds with
an emphatic no. The old dichotomy between fast, nimble, generally interesting endotherms and torpid,
clumsy, dull ectotherms, he says, is a false one. "People think we're making comments about stupid
lazy dinosaurs. We're not--we're not! We're just trying to come up with a real picture of them, and
that picture doesn't preclude their being mammal-like."

For inspiration, he suggests, we can turn to any number of interesting reptiles that manage to do things
that seem restricted to warm- blooded animals. Once a Komodo dragon bites an animal and infects the
wound with bacteria-infested saliva, it will track the escaping prey for hours as it dies. Pythons and pit
vipers are good parents to their children, attentively defending them from predators. And leatherback
turtles keep a high body temperature as they swim for thousands of miles through cold North Atlantic
waters, simply because their huge bulk traps the heat they produce or absorb from the sun while
basking at the surface.

A gigantic long-necked sauropod dinosaur might have been able to harness this kind of heat-trapping
well enough to stay warm 24 hours a day in many climates, and to travel long distances. Cold-blooded
bipedal carnivorous dinosaurs would still have been quite capable of swift running and terrifying leaps
on their prey. But instead of having endless stamina, they would have been ambushers, lurking through
forests and bursting upon their prey. "People think that if dinosaurs are not endotherms, they're just
lizards sitting on a rock, and I'm not saying that," says Ruben. In fact, he points out, the kind of life
laid out here compares relatively well with that of a lion, that supposed pinnacle of mammal hunters,
which sleeps most of the day and spends only a couple of hours hunting.

On the other hand, some of the graceful two-legged dinosaurs were clearly high-speed running
machines. "All I can think," says Owerkowicz, "is that they must have been active ectotherms. Maybe
they had dark muscle like tuna have--muscle that keeps the animals' internal organs warm by
contractions. I think it actually makes the dinosaurs much more interesting because it makes them
different. Their physiology is really not like anything we see today."

And that, after all, is what attracts us to these enormous extinct creatures of so long ago: they are
mysteries, and so it should not be surprising that they don't readily fit into either the warm-blooded or
cold-blooded slot. They were something all their own: they were dinosaurs.

- December 1996, Discover Magazine