Specimen · Tail plate & tail spike
Say it: STEG-oh-SORE-us · Greek for “roof lizard”
You are holding copies of two pieces of armour from the tail of a Stegosaurus: one of the bony plates that stood up along its back, and one of the four spikes at the end of its tail. Both were copied from a skeleton found by school students.
When
North American Stegosaurus fossils come from the Morrison Formation, rock laid down by rivers and floodplains between about 157 and 150 million years ago. Volcanic ash just above the Garden Park quarries, where this skeleton was found, has been dated to 152.3 million years.
Think about this: about 84 million years separate this Stegosaurus from Tyrannosaurus. Only 66 million separate T. rex from you. The two never met, whatever the films show.
Where
This skeleton came from Garden Park, just north of Cañon City, Colorado. The left globe shows today’s Earth with the main Stegosaurus sites. The right globe shows the same sites moved back to where they sat 150 million years ago, when North America was further south and much closer to Europe. Drag either globe.
How sure are we? The ancient map is a reconstruction, not a photo. Different plate models put these sites up to about 12° of latitude apart (more than 1,000 km). We use the PALEOMAP model for both the map and the pins, so they agree with each other.
Size
The skeleton below is the real 3D scan of the Denver Museum mount, at its true size. Your two pieces glow red. Next to it: an adult (1.75 m) and a 12-year-old (1.50 m). The ruler on the ground is in metres.
Skeleton: 3D scan of the Denver Museum of Nature & Science mount, true size. With skin: a simple low-poly model, scaled to the same height. It shows the general shape only; nobody knows what Stegosaurus skin looked like, or its colour.
These bones
In 1936 or 1937 (sources disagree), students from Cañon City High School on a field trip with their teacher Frank Kessler found bones weathering out of a hillside at Garden Park, Colorado. The Colorado Museum of Natural History (now the Denver Museum of Nature & Science) dug it up, paying students 25 cents an hour.
Preparator Phillip Reinheimer mounted it for display in 1938–39. In 1994–95 Kenneth Carpenter and Bryan Small took it apart and rebuilt it in a modern pose, turning to swing its tail at an Allosaurus. Steel was shaped around every bone and welded in place, so it can never be taken apart again.
In 1982, after a campaign by Colorado school children, Stegosaurus became the state fossil of Colorado. The governor’s order mentions this skeleton.
In 2018 Triebold Paleontology spent a week crawling inside the rib cage with an Artec 3D scanner to make the scan your replicas come from.
So the honest description of your replicas is: copies of the plate and spike as they appear on the Denver Museum skeleton. We have asked the museum, and this box will be updated when they answer.
A Stegosaurus had about 17–22 plates in two alternating rows, plus four tail spikes. The plates are small on the neck, biggest over the hips and the base of the tail, then get smaller again toward the spikes.
Diagram, not to scale: plate sizes follow “Sophie”, the most complete Stegosaurus known (London). One puzzle we have not solved: on Sophie, the last tail plates are under 20 cm tall, but on the Denver mount the last plate before the spikes is 42 cm. Either the Denver animal was bigger, or the mount places plates differently. It is a good reminder that a museum skeleton is partly an interpretation.
Known vs guessed
A lot of “dinosaur facts” are really good guesses. Scientists are careful about the difference, and you should be too. Every statement below has a label and a source.
Skeletons found with the plates still in place, including the 1992 Denver skeleton, show two rows, staggered left and right, not one row and not in pairs. The plates were not joined to the backbone; they sat in the skin.
Gilmore 1914, Smithsonian Bulletin 89; Carpenter 1998, Modern Geology 23; Maidment et al. 2015, PLOS ONE
Three ideas. Armour: unlikely; the plates are thin and full of blood-vessel channels, and they leave the sides unprotected. Heating and cooling: proposed in 1976 because of the grooves and channels for blood vessels; later studies found the plates were not built like a radiator, though they may have helped a little. Display: showing off to rivals or mates, or recognising its own species. Most scientists today lean toward display, but nobody can test it directly. Your plate shows the grooves that started the argument.
Farlow et al. 1976, Science; de Buffrénil et al. 1986, Paleobiology; Main et al. 2005, Paleobiology; Farlow, Hayashi & Tattersall 2010, Swiss Journal of Geosciences
Evidence: about 1 in 10 spikes studied has a broken and healed tip. An Allosaurus tail bone from Utah has a healed hole that a Stegosaurus spike fits. The 1992 Denver skeleton has a spike that broke and became infected while the animal was alive. Hypothesis: exactly how it swung the tail, and how often it fought.
McWhinney, Rothschild & Carpenter 2001; Carpenter et al. 2005, in The Carnivorous Dinosaurs; Carpenter 1998
Six skeletons found with their spikes together each had four. The idea of eight came from one jumbled specimen that may mix two animals. Old museum mounts pointed the spikes up; Carpenter, while rebuilding this very skeleton, showed from the spike bases and the 1992 tail that they pointed backwards and out to the sides.
Gilmore 1914; Carpenter 1998; Galton 2010, Swiss Journal of Geosciences
The spiky tail has a nickname from a 1982 Far Side cartoon by Gary Larson, in which a caveman names it “after the late Thag Simmons”. Palaeontologists liked it and started using it.
Evidence: the brain was very small for the body, roughly the size of a lemon. In 1881 Marsh noticed a big space in the spine above the hips and called it a “posterior braincase”, and the “second brain” story was born. Hypothesis: the space held a glycogen body, an energy store that birds also have there, not nerve tissue. What a glycogen body does is unknown even in birds.
Caspar et al. 2024, Anatomical Record; Giffin 1991, Modern Geology 16
The 1992 Denver skeleton has hundreds of small bony discs, 4–27 mm across, in the skin of its throat. Not every skeleton has them; they may have been lost before burial.
Carpenter 1998; Maidment et al. 2015
In 1978 Robert Bakker suggested Stegosaurus stood on its back legs and tail to reach high leaves. Carpenter argued the plates made the tail too stiff to lean on. A computer model of a smaller relative found it possible. No fossil shows it.
Bakker 1978, Nature; Carpenter 1998; Mallison 2010, Swiss Journal of Geosciences
Short front legs suggest it was not built for running. Stegosaur footprints in Spain give a walking speed of about 3 km/h, slower than you walk, and show several animals walking side by side. Those tracks were made by a stegosaur, not necessarily Stegosaurus itself.
Castanera et al. 2024, Scientific Reports
A 2015 study of a close relative suggested two plate shapes, wide and tall, belonging to the two sexes. A 2017 study found no statistical support for telling dinosaur sexes apart this way. Unresolved.
Saitta 2015, PLOS ONE; Mallon 2017, Paleobiology
Marsh named several species. The first name, S. armatus, was based on bones too poor to define it, so in 2013 the international naming commission made S. stenops, the Garden Park species, the official reference. Whether S. ungulatus is a separate species is still debated.
ICZN Opinion 2320, Bulletin of Zoological Nomenclature; Galton 2010; Maidment et al. 2008, J. Systematic Palaeontology
Bones from Batalha, Portugal, have been called “Stegosaurus cf. ungulatus” (“cf.” means “compare with”: similar, not certain). The material is a few bones and one plate, so this is a tentative identification.
Escaso et al. 2007, Naturwissenschaften; Costa & Mateus 2019, PLOS ONE
No Stegosaurus skin has been described. A close relative had small scales, and its plates had a smooth covering, probably keratin like your fingernails. The colour of the plates and body is unknown; every picture is an artist’s choice.
Christiansen & Tschopp 2010, Swiss Journal of Geosciences
Weight
Scientists either build a model of the body and multiply its volume by a guessed density, or measure the leg bones and compare with living animals. Leg-bone methods are tuned on adult mammals, so they overestimate a young, still-growing animal. When the best skeleton, “Sophie”, was checked both ways and corrected for her age, the answers agreed: about 1.6–2.2 tonnes.
| Estimate | Who, when | Which animal | Method |
|---|---|---|---|
| 1.8 t | Colbert 1962 | not stated | Sculpted scale model |
| 3.1 t | Alexander 1985 | not stated | Scale model |
| 2.2 t | Paul 1997 | USNM 4934 | Sculpted model |
| 2.5 t | Henderson 1999 | USNM 4934 | 3D computer model |
| 2.6 t | Seebacher 2001 | USNM 4934 | Equation from body shape |
| 2.7 t | Maidment et al. 2014 | USNM 4934 | 3D computer model |
| 3.8 t | Redelstorff & Sander 2009 | Sophie | Leg-bone equation |
| 7.0 t | Benson et al. 2014 | YPM 1853, a big individual | Leg-bone thickness |
| 1.6 t | Brassey et al. 2015 | Sophie | “Shrink-wrap” around the 3D skeleton |
| 1.8–2.2 t | Brassey et al. 2015 | Sophie | Leg bones, corrected for age |
Nobody has published a weight for the Denver animal itself. Its thigh bone is about 9 % longer than Sophie’s, so it was somewhat heavier. A fair answer for a classroom: roughly 1.5 to 3 tonnes, about the weight of a white rhinoceros or a large car. You may read “up to 10 tons” on some websites; we could find no study behind that number. Sources: Brassey et al. 2015 (free copy), Hayashi et al. 2012.
Sources & credits