Sir

An example of the visual deception practised by scientific illustrators seeking greater impact — as described by Julio M. Ottino in his Commentary (Nature 421, 474; 2003) — is the increasingly prevalent practice of adding foreground stars and background galaxies to images generated from numerical simulations of galactic collisions. Examples appear in National Geographic 203, 2, 2003; and on the Gadget website at http://www.mpa-garching.mpg.de/gadget. Another is one of my simulations, published unadulterated on the cover of Nature on 9 March 1989, which was later “improved” by the image-makers at NASA's Space Telescope Science Institute (http://hubblesite.org/newscenter/archive/2001/22/video/c) — without my consent.

Although one may agree that superimposed stars and galaxies add a bit of visual interest to the blankness of cyberspace, it is disturbing that these additions are entirely static. In reality, foreground stars hurtle past like snowflakes in a blizzard, and even background galaxies change noticeably over the hundreds of millions of years that are represented in the computer simulations. Static foreground stars and background galaxies undermine the basic idea of a dynamic Universe that these images and animations presumably attempt to convey.