What did Tyrannosaurus rex use its short arms for?

When a child looking at a T. rex skeleton asks why its arms are so small, the accurate answer is not one tidy explanation but two lists: what the bones show, and what researchers infer from them. By the end you will be able to say how long the arm bones actually are and which specimen the numbers come from, how many fingers T. rex had, which hypotheses have been proposed for the arms and their weak points, and what injured arm bones can tell us, and you will know how to read the arm pose of a mounted skeleton.

When, where and how big

Tyrannosaurus rex lived in the Late Cretaceous, about 68 to 66 million years ago, and its fossils come from the USA and Canada. It was named by H. F. Osborn in 1905.

Its size depends on the specimen and the method. The Natural History Museum in London gives a general length of 12.0 m and a weight of 7,000 kg without naming a specimen. SUE (FMNH PR2081), discovered in South Dakota in 1990, is 40.5 feet (about 12.3 m) long and 13 feet (about 4 m) tall at the hip. A 2016 Field Museum post says it probably weighed nine tons in life (about 8.2 t if these are US short tons). Hutchinson and colleagues (2011), working from 3D computer models of four large specimens, estimated adults at around 6,000 to 8,000 kg and SUE at perhaps 9,500 kg.

Measuring the arm bones: a 37 cm humerus and two fingers

To give an arm length in numbers, you first have to choose a specimen. Kevin Padian, a curator at the UC Museum of Paleontology in Berkeley, measured for a 2022 paper the mounted skeleton that stands outside the museum, inside the Valley Life Sciences Building. It is a cast based mainly on Museum of the Rockies specimen MOR 555, and Padian adds that reconstructions must be used with caution but that this specimen is relatively complete.

On that mount the humerus (upper arm bone) is 37 cm long, the ulna in the forearm 20 cm, and the first and second fingers 22 cm and 25 cm. The skull is 120 cm and the femur (thigh bone) 130 cm, so the humerus is less than a third the length of the skull and the femur about 3.5 times the humerus. Padian wrote that the inferences drawn from these measurements support earlier authors’ conclusion that the forelimbs could not reach the mouth, and that they could not reach each other or anything in front of the head.

The figure of about 1 m given for the whole arm is a general one, not a measurement of a named specimen. Padian puts the forelimb of a 12 m T. rex at only about one metre, and Steven Stanley of the University of Hawaii, in a 2017 conference abstract, also put the arms at about 1 m and argued they were not as tiny as often portrayed. A UC Berkeley news release likens the proportion to a 6-foot (about 183 cm) human with 5-inch (about 13 cm) arms; that is an illustration, not a measurement.

T. rex had two fingers. According to Padian, the oldest tyrannosauroids had three fingers. In some tyrannosauroids the third digit persists, but in T. rex it may be reduced to the point of losing its finger bones, and was perhaps sometimes lost altogether.

There is one thing to watch for at a museum. When the Field Museum remounted SUE in 2018, it corrected the furcula (wishbone) and adjusted the shoulder blades, which lowered the arms and brought them forward a little, and the museum says they now look as though they could touch each other. That rests on a different specimen and a different kind of evidence from Padian’s conclusion, so the two cannot be set directly against each other. The people who mount a skeleton decide where the arms go, so a display pose alone says little about what the living animal could actually do.

DINO NOTES

Bone lengths measured on one mount

The UC Berkeley mount, a cast based mainly on MOR 555 (Padian 2022)

• Skull

  120 cm

  • 60 cm wide at the jaw joints

• Femur (thigh bone)

  130 cm

  • About 3.5 times the humerus

• Humerus and ulna (forearm)

  37 cm · 20 cm

  • The humerus is under a third of the skull

• Two fingers

  22 cm, 25 cm

  • Digit I 22 cm, digit II 25 cm
  • Digit III may have lost its phalanges

Measured on a restored cast, not on original bones. Padian concludes these arms could reach neither the mouth nor each other.

What were they for? The hypotheses and their weak points

How T. rex used its arms is still debated. These are the ideas proposed so far and the weak points raised against them.

A prop for getting up. B. H. Newman (1970) published a figure of the animal rising from rest, raising its head and straightening its forelimbs, and some scientists think the arms, which were attached to strong muscles, helped T. rex stand up after a fall. Padian points out that Newman proposed this without any mechanical analysis, and that animals with larger forelimbs would have made a better anchor.

Holding prey or a mate. Osborn (1906) suggested a grasping organ during mating, but Padian objects that the copulatory organs of vertebrates are at the pelvis, not the chest. Padian reports that Carpenter and Smith (2001) reconstructed the arms as able to lift some 400 lbs (nearly 200 kg) and saw them as holding prey; he replies that the jaws would have been buried in the prey before the arms could reach it, and calls the idea inviable and untestable. The Natural History Museum says young T. rex had arms more in proportion to their bodies that might have helped grab food, but Padian says there is not enough evidence to test this.

Slashing at close range. Stanley (2017) cited six traits, including 8 to 10 cm sickle-shaped claws and a large coracoid (a shoulder bone) showing the arms were very strong, to argue they were adapted for slashing at close quarters. This is a conference abstract, not a peer-reviewed paper.

Shrinking from disuse. Proposed causes of the shortening include natural selection, sexual selection, energetic compensation, quirks of growth and loss through disuse, but Padian argues that disuse by itself does not explain the reduction.

Keeping out of the way when eating together. This is Padian’s own hypothesis. Several quarry sites preserve adult and juvenile tyrannosaurs together, which has raised the possibility that they hunted in groups. Padian proposes that the huge skull and jaws did the predatory work, and that shorter arms were selected so they would not be bitten, torn off or infected when several animals fed on one carcass. He also set out ways to test it: if fewer bite marks were found on the reduced arms, that could be a sign the reduction worked, and younger animals should not show the same pattern of damage. Even so, Padian said firmly establishing the hypothesis may never be possible.

In 2026 researchers at UCL and the University of Cambridge (Proceedings of the Royal Society B) looked at a pattern rather than a use. Across 82 theropod species, reduced forelimbs were more strongly linked to skull robustness, scored partly by how tightly the bones of the head were connected, than to skull size or overall body size, and T. rex scored highest on that measure. The team reads this as the head taking over from the arms as the method of attack, while stating that a study of correlations cannot establish cause and effect.

DINO NOTES

Ideas about the arms and how they stand

Who proposed each idea, and what was said against it

• A prop for getting up

  • Proposed by Newman (1970)
  • No mechanical analysis (Padian)

• Holding prey or a mate

  • Mate: Osborn (1906)
  • Mating organs are pelvic (Padian)
  • Prey: Carpenter and Smith (2001)
  • The jaws get there first (Padian)

• Slashing at close range

  • Stanley (2017), conference abstract
  • 8-10 cm sickle-shaped claws, etc.
  • Not a peer-reviewed paper

• Shrank from disuse

  • One of several proposed causes
  • Not enough on its own (Padian)

• Out of the way at a shared carcass

  • Padian's hypothesis (2022)
  • Test: fewer bite marks on the arms?
  • May never be firmly established

A 2026 study of 82 theropod species: short arms are more strongly linked to skull robustness than to skull or body size. A correlation, not a proven cause.

What injured arm bones tell us

What SUE’s small arms did is still unknown, but its upper arm bone, the humerus, shows an injury. According to the abstract of Brochu’s 2003 monograph on SUE, lesions on the right scapulocoracoid (shoulder blade) and humerus coincide with fractured ribs on the right, which may indicate a single trauma to the right side of the body. In 2016 SUE’s right forelimb was removed and scanned with micro-CT to see the blood vessels and muscle attachments inside the bone and learn how the arm was used.

Senter and Juengst (2016), in a paper on forelimb bone disorders in Dilophosaurus, summarise earlier reports in their introduction. They list one T. rex with, among other things, a fractured furcula, a possible tendon avulsion on the right humerus and a deep pit on the right first metacarpal that may be due to gout. No specimen number is given, so it is not clear whether this is SUE. The paper also mentions a different specimen whose left humerus shows extensive periostitis, apparently from a tendon avulsion. It says that in non-avian theropods the shoulder girdle and forelimb are frequently afflicted, which suggests vigorous use of the forelimbs. Padian, for his part, reads reported humerus fractures as a possible sign that the bone was poorly suited to bearing forces such as the weight of the body. Injuries may be a clue that the arms were used, but they do not say for what.

Not every mark on an arm bone is an injury. Longrich and colleagues (2010) found in museum collections four T. rex specimens bearing tooth marks of large carnivorous dinosaurs, one of them a left humerus with a series of scores up to 36 mm long on its back surface. They argue these traces come from T. rex feeding on its own kind, not from fighting.

DINO NOTES

What fossils can and cannot tell

Sorting the evidence on T. rex arms

• Can be measured

  • Bone lengths and finger count
  • Muscle attachment sites; inner structure by CT

• Signs of injury and disease

  • SUE: lesions on the right humerus
  • Reports gathered in 2016: possible tendon avulsions, possible gout

• Signs of being eaten

  • Tooth scores on a humerus
  • Read as feeding by its own kind (2010)

• Not shown directly

  • What the arms were used for
  • Arm pose: people set it on a mount

So the uses stay hypotheses. They can be tested against bone lengths, where injuries sit and comparisons across species.

Answering the question

As the Natural History Museum puts it, the arms were likely used in multiple ways, but proving that more than 66 million years later is difficult. So the answer can come in two parts. What the bones confirm: the arms were short for the body, there were two fingers, and some arm bones carry marks of injury or of being eaten. What the arms were for is still a matter of hypotheses. Working out reach from bone lengths and comparing many species have already been done; comparing bite marks on different parts of the body is a method Padian has proposed.

At the museum, it is worth counting together how many fingers the mounted skeleton has and how short its arm bones are next to the skull and legs, and adding that the arm pose is a judgement made by the people who mounted it.