In the Beginning Was the Tooth

Discover, January 31, 1993

When paleontologists dissolve a chunk of limestone in acid, they often find hundreds or even thousands of cone-shaped, toothlike objects called conodonts (which is Greek for cone-shaped, toothlike object). In limestones that were laid down on the ocean floor between 515 million and 208 million years ago–that is, from the Late Cambrian through the Triassic periods–conodonts are all but ubiquitous. But what are they, and who did they belong to? Paleontologists have debated the question for well over a century, and at one time or another they have pinned conodonts on all manner of animals and plants.

This past year, however, the question may have been settled at last. Conodonts are indeed teeth, it seems. What’s more, they are teeth that belonged to vertebrates, which means the first vertebrates, our ancestors, lived at least 515 million years ago.

Vertebrates belong to a phylum of animals known as chordates, most of which are wormlike beasts but all of which have a stiff rod of tissue that protects the nerve running down their back. In vertebrates that rod, known as the notochord, evolved into a spine. But you don’t need vertebrae to be a vertebrate. What sets us apart on the most fundamental level is the ability to make bone. Other animals can enclose their bodies in shells by secreting minerals–usually calcium carbonate. Only vertebrates grow a framework inside their bodies made of bone, which consists of living cells embedded in a matrix of collagen fibers (the stuff of cartilage) and calcium phosphate. Vertebrate cells can also produce a shell-like substance called enamel, which coats teeth.

Until this past year the earliest known vertebrates were a group of 475-million-year-old armored fish. They had hard plates of bone underneath their skin, but they had no backbones. That led researchers to believe that bone evolved so that vertebrates could imitate shellfish and protect themselves from predators.

The new evidence from conodonts contradicts that notion. The mystery surrounding the original owners of these mysterious objects began to unravel in 1982, when Euan Clarkson of the University of Edinburgh discovered a fossil, lying unexamined in a museum drawer, that preserved both conodonts and the impression of a soft skeletonless body. The animal looked like a small eel, measuring two inches long, and the conodonts, about the size of sand grains, were in its mouth. The body showed evidence of a notochord, which suggested to some that the conodont animal was at least a chordate. But other researchers disputed the notion, let alone the idea that the conodont animals might be vertebrates; a recent edition of an authoritative textbook continued to put them in a phylum all their own.

Last May, though, Ivan Sansom, of the University of Birmingham in England, and his colleagues reported that conodonts were teeth and that they were made of bone–meaning that the conodont animal was a vertebrate. Neither point was obvious from the fact that conodonts had been found in the animal’s mouth; the cones might just have been internal mouth supports, or they might have acted as teeth but not actually been made of bone. (Some nonvertebrate chordates, for example, grow teeth from keratin, the protein in human fingernails.)

But the toothy and bony nature of conodonts became apparent when Sansom examined nearly a hundred of them, representing 23 separate species, under an electron microscope. In the past, paleontologists had prepared conodonts for the microscope as they would prepare a fossil shell, etching a cross section of the fossil with a strong hydrochloric acid solution to bring out its crystalline structure. Sansom, in contrast, used a much milder solution of chromium sulfate–the same one used by dental researchers to study teeth. “We were sitting there, and these images appeared that were quite spectacular,” remembers Sansom. “It was very, very clear we were looking at enamel, bone, and cartilage.”

Those are the elements of a vertebrate’s tooth: the bony core is covered by enamel, and the cartilage helps root the tooth in the mouth. Under the electron microscope, bunches of enamel crystals were organized into long rows, while the fossilized bone had spidery holes where the cells used to be, and the cartilage looked like little spongy globs. “It came as quite a shock,” says Sansom. “The structures were actually clearer in the conodonts than in human teeth.” The clear-cut patterns had been eaten away, however, when paleontologists had used stronger acids.

If Sansom is right, vertebrates are as old as the oldest conodonts–that is, at least 515 million years old–and they evolved bony teeth while remaining otherwise squishy. Only later did they adapt bone for other purposes, such as armor or a skeleton. It’s a revolutionary idea, and one that moves the conodont animals, once an obscure enigma, into the spotlight of evolution. “They were probably a trivial part of the ecological system,” says Sansom, “but they were part of the lineage that gave rise to us.”