As a professional scientific illustrator I feel compelled to respond to Julio Ottino's Commentary “Is a picture worth 1,000 words?” (Nature 421, 474–476; 2003). I believe that Ottino's criticisms of scientific illustration are founded on an incorrect understanding of the field.

Galileo's drawings can't be compared with magazine covers: they are two unrelated types of illustration. It is incorrect to conclude from such a comparison that scientific disparity exists between them because the magazine covers “are left in the hands of artists and illustrators” — this artwork was intended to enhance editorial material rather than to illustrate research.

Such conceptual illustrations are designed to pose questions. Their use on the cover of a science magazine offers the promise of articles that inform these questions. The cover art of the 30 January 2003 issue of Nature (see figure) and the related News and Views and Letter (Nature 421, 489–490; 2003 & Nature 421, 530–533; 2003) follow this convention. The image does not illustrate the research itself; that is not its intended application. Furthermore, the choice of digital medium, whether used by the hand of a scientist or the hand of an artist, has no bearing on this question.

Scientific illustration follows a different mandate, and it can often be found within the pages of the very magazines under discussion. These drawings outline structure and clarify detail, as required by the subject and requested by the researcher. Because they communicate subtleties and eliminate the ambiguities of language, scientific illustrations are an important, often necessary, element in precise communication (see The Guild Handbook of Scientific Illustration, edited by E. R. S. Hodges; Van Nostrand Reinhold, 1989).

Scientific illustration is a clearly defined field that benefits from active collaboration between scientist and illustrator. Using their professional observational skills, scientific illustrators strive to render the most accurate representation of their subject. It is, by definition, art in the service of science. The act of drawing is, in essence, the act of editing. Complaints about omitted details miss this important point. Scientific illustrators are trained to eliminate non-essential information. The twisted stem of a dried plant is smoothed out. The broken edge of a fossil bone is repaired. Cracks and discoloration may be removed. These subjects are thus rendered in a way chosen to amplify those details that require emphasis.

Scientific illustrations, even conceptual cover art, should be as accurate as possible. However, Ottino's proposal to establish rules governing the use of realistic rendering techniques is superfluous. Professional standards are already in place for scientific illustration. Magazine editors recognize that their educated readership can distinguish between a beautifully rendered concept and the current state of scientific research. Scientific illustrations exist within this context. They communicate with and within conventions that reach back in time from this issue of Nature to the pages of Galileo's notebooks.