University of Bristol

Dinosaur Gait and Estimates of Speeds


Estimates of Speeds of dinosaurs


Gait describes the pattern and synchronony of footfalls made by an animal. Gait often changes with increasing speed to maximise energetic expenditure. The study of gait is well documented for many modern species including the horse. Gait can be studied in detail by using high speed photography. This is obviously not an option for examining gaits of non-avian dinosaurs.

Instead a predictive method must be used by examining trackways and skeletal reconstructions. However, we can never know the relative phase (the timing of a footfall in relation to the other feet) or duty factor (the time the foot remains on the substrate). This creates problems for interpretations of trackways, especially those of quadrupedal tetrapods. Gaits of extinct animals are classified as walk, trot and run. Although quite arbitrary, this classification does have advantages:

Gaits are therefore described by stride length, which can be measured directly from trackways, and by the relative size of the animals. This relationship between size and stride length is termed relative stride length "l /h, where l represents length of stride and hrepresents the trackmaker's height at the hip" (Thulborn 1989).

Gaits are generally believed be l /h < 2.0 for walking, l /h = 2.0 for a trot and l /h > 2.9 for a run (Thulborn, 1984). Hip height can be estimated using a number of methods; two are described below.

Geometric Method

Drawing by M McLeod after Avnimelech (1966) in Thulborn (1989)

Using the geometric method illustrated above, suceeding footprints ('X' and 'Y') are assumed to be the base of an isosceles triangle. 'H' represents hip height.

The morphometric method was suggested by Alexander (1976). Alexander assumed from estimates of foot joint features that hip height was four times the length of the hind footprint for dinosaurs, whether bipedal or quadrupedal. This can be regarded as a crude, but useful ratio.

Return to top of page

Estimates of Speeds

Since it is likely that dinosaurs spent the majority of their time walking, rather than running we are more likely to find trackways of walking dinosaurs. When we do find a running dinosaur track, we cannot know whether it represents the top speed of that individual or species.

In 1976, R.McNeil Alexander published an exciting paper in Natureentitled "Estimates of speeds in dinosaurs" (Alexander, 1976). He used a number of trackways for his calculations of stride length. Some represented bipedal dinosaurs, others quadrupedal sauropods.

Alexander used mathematical principles to estimate the interaction of proportion with speed and gait. Alexander wished to produce an equation that could be applied to animals of many different sizes. This meant he needed a "non-dimensional parameter to serve as a criterion for physical similarity" (Alexander, 1976). To this end, Alexander used the Reynolds number, which relates to the interaction of inertia and viscous forces, and the Froude number. The Froude number relates to interaction between inertia and gravity. He used the following equation to estimate speed:

u a 0.25g-0.5d1.67h-1.17

where u = velocity; g= the acceleration of free fall; d = stride length; h= hip height.

The fastest dinosaur that he recorded was moving at 3.6 ms-1. Alexander suggested this was more characteristic of the traditional lumbering image of dinosaurs.

Since the publication of this paper a number of other trackways have been examined. The equation developed by Alexander has undergone various corrections and modifications by Alexander and other authors (Thulborn, 1981; Farlow, 1981; Russell & Beland, 1976). Top speeds reported by authors have suggested some large theropods, such as Tyrannosaurusmay have reached up to 10 to 15 ms-1.

Alexander (1996) argued that based on the bone dimensions of Tyrannosaurusit is unlikely they could have travelled at more than 8ms-1.

Return to top of page


Alexander, R.M. (1976) Estimates of speeds of dinosaurs. Nature.Vol.261, pp.129-130.

Alexander, R.M. (1996) Tyrannosaurus on the run. Nature.Vol.379, pp.121.

Avnimelech, M.A. (1966) Dinosaur tracks in the Judean Hills. Proceedings of the Israel Academy of Sciences and Humanities,Section of Sciences. Vol.1, pp.1-19.

Farlow, J.O. (1981) Estimates of dinosaur speeds from a new trackway site in Texas. Nature. Vol.294, pp.747-748.

Russell, D.A., Beland, P. (1976) Running Dinosaurs. Nature. Vol.264, pp.486.

Thulborn, R.A. (1981) Estimated spped of a giant bipedal dinosaur. Nature. Vol.292, pp.273-274.

Thulborn, R.A. (1984) Prefered gaits of bipdeal dinosaurs. Alcheringa Vol.8, pp.243-252.

Thulborn, R.A. (1989) The Gaits of dinosaurs. In Dinosaur Tracks and Traces (eds. D.D.Gillette, M.G.Lockley), pp.39-50. Cambridge University Press, Cambridge.



Back to the Dinosaur tracks and trails home page
Back to The Palaeofiles home page
Back to Bristol Palaeobiology Group home page