TAPHONOMY
The Trilobite’s Calcite Eyes
11 thoracic segments, calcite lenses
Trilobites looked out through calcite. Each lens of a phacopid eye was a single crystal set with its optical axis perpendicular to the surface, so head-on light is not doubled. In 1975 Clarkson and Levi-Setti found doublet lenses that correct spherical aberration, and Niles Eldredge’s counts of lens files in this species helped found punctuated equilibrium.
TAPHONOMY
The Ammonitic Suture
lobes and saddles both frilled
Where each septum met the shell it left a suture, visible only once the shell is gone. Sutures sort the ammonoids: plain zigzags in Devonian to Permian goniatites, serrated lobes in Triassic ceratites, fern-like frills in the ammonites of the Jurassic and Cretaceous. Why they grew so elaborate is still argued.
TAPHONOMY
A Gnat in Baltic Amber
two wings, two halteres
Baltic amber is the resin of Eocene forests, marked by star-shaped oak hairs and a milky film on one side of many insects. It keeps bristles and wing veins, but not DNA: a 242-base stretch has a half-life of 521 years. This insect has two wings and two halteres, so it is a fly, a fungus gnat.
TAPHONOMY
Archaeopteryx, the Berlin Specimen
teeth, three claws, a bony tail
Two years after the Origin of Species, the Solnhofen quarries yielded a feather, then a skeleton with feathered wings, teeth, clawed fingers and a long bony tail. The Berlin specimen, found in 1874 or 1875, was the first with a complete head. Like almost every articulated Archaeopteryx, it lies in the death pose, head thrown back.
TAPHONOMY
Tyrannosaurus rex
D-shaped front teeth, four a side
Barnum Brown dug the first Tyrannosaurus rex from the Hell Creek Formation of Montana in 1902, and Osborn named it in 1905. Its skull is a row of archosaur landmarks: nostril, antorbital fenestra, keyhole-shaped eye socket, lower temporal opening. The four small front teeth on each side of the upper jaw are D-shaped in section, a tyrannosaurid signature.
NOT A DINOSAUR
Dunkleosteus
no teeth: self-sharpening bone
Dunkleosteus had no teeth. Its jaws were bony plates worn against each other into self-sharpening blades, rising at the front into fang-like points. In 2007 Anderson and Westneat modeled the skull as a four-bar linkage and found jaws that opened fast and bit hard. A typical adult was about 3.4 meters long, not the old five to ten.
NOT A DINOSAUR
Tiktaalik
a fish with a neck and a wrist
Shubin, Daeschler and Jenkins picked Ellesmere Island because its rocks, Late Devonian stream deposits about 375 million years old, were the right age to hold something between fish and the first four-legged animals. They found Tiktaalik in 2004: gills, scales and fin rays, but a flat head, a neck, and a fin with a wrist.
NOT A DINOSAUR
Smilodon
not a dinosaur, and not a tiger
The saber-toothed cat of the Rancho La Brea asphalt seeps is neither a tiger nor a dinosaur. It is a machairodont, from an extinct branch of the cat family, with flattened, finely serrated upper canines and a jaw that dropped far wider than a lion’s. The asphalt stained its bones brown-black. It is California’s state fossil.
NOT A DINOSAUR
Dimetrodon
a synapsid: our side of the tree
Dimetrodon is in nearly every box of toy dinosaurs, but it lived in the Early Permian and died out some forty million years before the first dinosaur. It is a synapsid, with one opening in the skull behind the eye, on the line that led to mammals. Cope named it in 1878 for its two measures of teeth.
NOT A DINOSAUR
Pteranodon
one finger holds the wing
Marsh named it in 1876 from the Niobrara Chalk of Kansas: wing without tooth. Pterosaurs were the first vertebrates to fly under their own power, and they are not dinosaurs. The wing gives them away: a bat uses four fingers, a bird a fused hand, but a pterosaur hung its whole outer wing from one long fourth finger.
STEM GROUP
Dickinsonia
an animal by its sterols, 2018
Dickinsonia lived about 558 million years ago, before the Cambrian, and for decades nobody could agree what it was. In 2018 Ilya Bobrovskiy and colleagues extracted molecules from its thin organic film and found almost nothing but cholesteroids, the sterols of animals. Chains of faint impressions show that it rested, apparently fed on the microbial mat, then moved on.
STEM GROUP
Marrella
two pairs of head spines, 26 segments
Marrella splendens is the commonest animal in Walcott’s quarry of the Burgess Shale: Walcott found it in 1909. Harry Whittington showed in 1971 that it was neither trilobite nor crustacean. In 1998 Andrew Parker reported diffraction gratings on its surface, so in life its spines would have flashed iridescent color.
STEM GROUP
Anomalocaris
its claw was first named as a shrimp
Its kind was named three times before anyone saw a whole one: a spiny appendage described as a shrimp in 1892, and in 1911 a round, plated mouth named as a jellyfish and a body as a sea cucumber. Harry Whittington’s work in the early 1980s showed such parts belonged to large swimming predators, the radiodonts.
STEM GROUP
Opabinia
five eyes and a nozzle
When Harry Whittington showed an early reconstruction to a meeting of paleontologists in 1972, the room burst out laughing. Opabinia has five eyes, a flexible nozzle about four times the head’s length ending in a spiny claw, and fifteen segments with gill-bearing flaps. Graham Budd placed it on the arthropod stem in 1996, near the radiodonts.
STEM GROUP
Hallucigenia
flipped in 1991, head found in 2015
Simon Conway Morris named it in 1977 for its dreamlike look and stood it on seven pairs of spines. In 1991 Lars Ramsköld and Hou Xianguang turned it over: the spines were on its back and the tentacles were legs. In 2015 Martin Smith and Jean-Bernard Caron found its head, at the other end.
DEAD CLADE WALKING
Lystrosaurus
two tusks, no other teeth
Lystrosaurus, the shovel lizard, was a pig-sized, beaked plant-eater with only two teeth, a pair of tusks. It crossed the end-Permian extinction and flooded the emptied world: in the earliest Triassic of the Karoo it makes up most of the vertebrate fossils. Found in Antarctica in 1969, it helped show the southern continents were once joined.
DEAD CLADE WALKING
Shocked Quartz
planar features: about 10 GPa and up
A shock from an impact leaves a signature no volcano can: thin, dead-straight lamellae across a quartz grain, in several sets, formed from about 10 GPa. In 1984 Bruce Bohor and colleagues found such grains in the boundary clay in Montana. With the iridium they pointed to an impact, and Chicxulub, about 180 km across, was identified in 1991.
DEAD CLADE WALKING
The Iridium Anomaly
about 30 times background at Gubbio
Walter Alvarez wanted to know how long the centimeter of clay at Gubbio had taken to settle, and his father Luis suggested measuring iridium. With Frank Asaro and Helen Michel they found 0.3 parts per billion below it and 9.1 in its red upper half, more than thirty times higher. Above it, the excess halves every 4.6 cm.
DEAD CLADE WALKING
The Big Five
Raup and Sepkoski, 1982
In 1982 David Raup and Jack Sepkoski picked out five extinctions above the background in the marine fossil record. On the International Commission on Stratigraphy’s chart, version 2024/12, they fall 18, 31, 53, 63 and 88 percent of the way along the Phanerozoic, where the notches are cut. The end-Permian is the worst by far. The relief is a schematic of Sepkoski’s curve.
DEAD CLADE WALKING
Mammoth and Mastodon
Cuvier, 1796: extinction is real
In 1796 Georges Cuvier showed that the mammoth of Siberia and the Ohio animal differed from both living elephants. Animals that large could hardly be hiding, so they were gone: extinction was real. The teeth tell it at a glance: a mammoth’s grinding mill of thin enamel plates for grass, a mastodon’s paired cusps for browsing.
NO VESTIGE
Superposition and Index Fossils
oldest at the bottom: Steno, 1669
Nicolaus Steno set out the rule in 1669: in undisturbed layers the lowest is the oldest. William Smith, a canal surveyor, found that each layer also carries its own fossils in a fixed order. The best index fossils lived briefly, spread widely and are easy to tell apart, and they still do most day-to-day correlation.
NO VESTIGE
Zircon, the Timekeeper
Jack Hills, 2001: 4.404 Ga
Zircon takes uranium into its lattice as it grows and all but excludes lead, so its clock starts near zero, and its closure temperature for lead is above 900 °C. In 2001 a Jack Hills zircon dated at 4.404 billion years became the oldest known piece of the Earth: 0.98 atoms of lead-206 for every atom of uranium-238 left.
NO VESTIGE
Stromatolites
convex side up: the way-up test
A stromatolite is a rock built by microbes: mats trap sediment and help carbonate precipitate, each new mat over the last. Probable stromatolites in the Pilbara are 3.48 billion years old. Their laminae bulge upward, so a mapper who finds them upside down knows the beds are overturned.
NO VESTIGE
Hutton’s Unconformity
Siccar Point, 1788: a 65-million-year gap
In 1788 James Hutton took Playfair and Hall by boat to Siccar Point, where Silurian graywackes stand on end, worn flat and buried under Old Red Sandstone. He read the contact as sea floor, uplift, erosion and burial again, at the slow rates of his own day. The gap is about 65 million years.
NO VESTIGE
The U-Pb Concordia
half-lives 4.468 and 0.704 billion years
Uranium carries two clocks: uranium-238 decays to lead-206 with a half-life of 4.468 billion years, uranium-235 to lead-207 with 0.704. Plot one ratio against the other and every undisturbed crystal falls on Wetherill’s 1956 concordia. A crystal that lost lead slides off along a chord whose ends give both its age and the disturbance.