Unfortunately, Haeckel was overzealous. When we compared his drawings with real embryos, we found that he showed many details incorrectly. He did not show significant differences between species, even though his theories allowed for embryonic variation. For example, we found variations in embryonic size, external form, and segment number which he did not show (1). This does not negate Darwinian evolution. On the contrary, the mixture of similarities and differences among vertebrate embryos reflects evolutionary change in developmental mechanisms inherited from a com mon ancestor (5).

Haeckel's drawings are used in many modern textbooks, but not always as primary evidence for evolution. In Molecular Biology of the Cell (6), the drawings are used mainly to support hypotheses about the stages of development acted on by natural selection. It is only in this limited context that we have reservations about the implications of the drawings. Thus, certain "phylotypic" embryonic stages, which Haeckel showed as identical, may in fact be significant targets for natural selection.

We are not the first to question the drawings. Haeckel's past accusers included His (Leipzig University), Rütimeyer (Basel University), and Brass (leader of the Keplerbund group of Protestant scientists). However, these critics did not give persuasive evidence in support of their arguments. We therefore show here a more accurate representation of vertebrate embryos at three arbitrary stages, including the approximate stage (Fig. 1, column three), which Haeckel showed to be identical. We suggest that Haeckel was right to show increasing difference between species as they develop. He was also right to show strong similarities between his earliest embryos of humans and other eutherian mammals (for example, the cat and the bat; Fig. 1, column three). However, he was wrong to imply that there is virtually no evolutionary change in early embryos in the vertebrates (see variations, Fig. 1, column three).

These conclusions are supported in part by comparisons of developmental timing in different vertebrates (7). This work indicates a strong correlation between embryonic developmental sequences in humans and other eutherian mammals, but weak correlation between humans and some "lower" vertebrates. Haeckel's inaccuracies damage his credibility, but they do not invalidate the mass of published evidence for Darwinian evolution. Ironically, had Haeckel drawn the embryos accurately, his first two valid points in favor of evolution would have been better demonstrated.

Michael K. Richardson
Department of Anatomy and
Developmental Biology,
St. George's Hospital Medical School,
London SW17 ORE, United Kingdom
E-mail: m.richardson@sghms.ac.uk
JamesHanken
Department of Environmental,
Population, and Organismic Biology,
University of Colorado, Boulder, CO 80309-0334, USA.
Lynne Selwood
Department of Zoology,
La Trobe University,
Bundoora, Victoria, 3083, Australia
Glenda M. Wright
Department of Anatomy and Physiology,
Atlantic Veterinary College,
University of Prince Edward Island,
Charlottetown, Prince Edward Island,
C1A 4P3, Canada
Robert J. Richards
Fishbein Center For History of Science,
University of Chicago,
Chicago, IL 60645, USA
Claude Pieau
Département Dynamique du Génome et Evolution,
Institut Jacques Monod, 2, Place Jussieu,
75251 Paris, Cedex 05, France
Albert Raynaud
Laboratoire de Zoologie,
Université Paul Sabatier,
31062, Toulouse Cedex, France

References and Notes

1. M. K. Richardson et al., Anat. Embryol. 196, 91 (1997).

2. "Firing Line with William Buckley," Public Broadcasting System (USA) (13 December 1997).

3. W.H. Rusch, Creat. Res. Soc. Ann. 6, 27 (1969).

4. J. M. W. Slack, P. W. H. Holland, C. F. Graham, Nature 361, 490 (1993).

5. R. Raff, The Shape of Life (Univ. of Chicago Press, Chicago, IL, 1996).

6. B. Alberts et al., Molecular Biology of the Cell (Garland, New York, ed. 3, 1994).

7. M. K. Richardson, Dev. Biol. 172, 412 (1995); and M. Coates, in preparation.

8. F. Verbeek, J. Bluemink, J. Narraway, and other staff of the Netherlands Institute for Developmental Biology provided assistance. Supported by the Wellcome Trust.