Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Insect behaviour

Motion camouflage in dragonflies


Most animals can skilfully conceal themselves when stationary1, but they may become apparent as soon as they move. Here we use stereo cameras to reconstruct the movements in three dimensions of dragonflies (Hemianax papuensis), and show that these insects actively use motion camouflage to disguise themselves as stationary during territorial aerial manoeuvres. Deployment of this sophisticated technique by the oldest airborne predator tricks the victim's retina into perceiving the stalker as stationary even while it darts about in pursuit.


Optic flow — the apparent movement of objects as perceived using the retina — is a primary cue for detecting predators and prey. Predatory animals attacking stationary prey generally attempt to conceal their presence by approaching very slowly, sometimes relying on a camouflaging background. Active motion camouflage has been proposed as a strategy by which a predator can conceal its movements while shadowing or attacking highly manoeuvrable prey2.

Motion camouflage can be achieved if one animal (the shadower) moves in such a way as to produce the same image motion on the retina of another animal (the shadowee) as would a stationary object in the environment. We reconstructed 15 three-dimensional flight trajectories of interactions between conspecific dragonflies, of which six showed clear evidence of active motion camouflage.

Figure 1a shows an interaction between two male dragonflies: the lines connecting the heads of the shadower (blue dots) and shadowee (red dots) intersect in a small region behind the shadower. The centre of the shaded sphere represents the best estimate of the common point of intersection, and the radius of the sphere represents the uncertainty of this estimate.

Figure 1: Examples of active motion camouflage.

a, Three-dimensional reconstruction of a territorial interaction involving two male dragonflies. The reconstruction was achieved by using calibrated6 and synchronized (to within 10 μs) stereo cameras. Positions of the heads of the shadower (blue dots) and shadowee (red dots) are connected by lines at corresponding 20-ms steps. The connecting lines converge behind the shadower; the point of convergence indicates the apparent position of the shadower to the shadowee. The shaded sphere (radius, 11.3 mm) represents the uncertainty of the 'best' estimate of the intersection point. b, Angular-velocity profile produced by the shadowing dragonfly in the shadowee's eye (filled circles) compared with that produced by a virtual stationary object at the intersection point (hollow circles). c, Another example of motion camouflage. The connecting lines converge in front of the shadower during frames 1–8, after which the shadower begins to imitate an infinitely distant object. The radius of the intersection sphere is 8.5 mm.

If motion camouflage were perfect, all lines would intersect at a single point, and the sphere of uncertainty would have zero radius. In this example, the radius of the intersection sphere is 11.3 mm. Given the ±10-mm accuracy in measuring position at a filming distance of 1.5 m, we may say that the lines intersect almost at a common point. This indicates that the shadower is emulating the trajectory of the projection of the fixed intersection point on the shadowee's retina. In other words, as far as the shadowee is concerned, the shadower is indistinguishable from a stationary object positioned at the intersection point.

The effectiveness of the strategy of motion camouflage is illustrated by a comparison of the apparent motion of the shadower on the shadowee's retina with that produced by a virtual, stationary object positioned at the estimated intersection point (Fig. 1b). The two angular-velocity profiles are very similar, indicating that the strategy is likely to be effective in concealing the motion of the shadower.

In other male–male interactions, we observed variations in motion camouflage (Fig. 1c). During the first eight frames of filming, the lines connecting the two insects intersect at a small region between them; the shadower is therefore emulating a fixed object in front of it. Here the shadowing animal is not flying towards the shadowee, but is moving in the opposite direction. This is evidence that the flight pattern of the shadower is motivated by motion camouflage and is not an artefact or by-product of chasing per se2,3. After frame nine, the intersection point shifts to infinity — that is, the shadower imitates an object at infinity. This example reveals that a shadower can combine two different types of motion camouflage in a single episode.

Active motion camouflage requires the shadower to move in such a way that it imitates the trajectory of a fixed object on the retina of the shadowee by precise flight control and positional sensing4, although exactly how the shadower achieves this is unclear2. Given the high sensitivity of the insect visual system to movement5, motion camouflage is likely to be the key to the success of such manoeuvres.


  1. 1

    Curio, E. The Ethology of Predation (Springer, Berlin, 1976).

    Book  Google Scholar 

  2. 2

    Srinivasan, M. V. & Davey, M. Proc. R. Soc. Lond. B 259, 19–25 (1995).

    ADS  Article  Google Scholar 

  3. 3

    Collett, T. S. & Land, M. F. J. Comp. Physiol. 99, 1–66 (1975).

    Article  Google Scholar 

  4. 4

    US Office of Naval Research Insect Vision Based Seekers and Controllers for Guided Weapons (contract N68936-1-2-001).

  5. 5

    Srinivasan, M. V. in Visual Motion and its Role in the Stabilization of Gaze (eds Miles, F. A. & Wallman, J.) 139–155 (Elsevier, Amsterdam,1993).

    Google Scholar 

  6. 6

    Tsai, R. Y. IEEE J. Robot. Automat. 3, 323–344 (1987).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Akiko Mizutani.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Mizutani, A., Chahl, J. & Srinivasan, M. Motion camouflage in dragonflies. Nature 423, 604 (2003).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing