Posted June 1, 2005
The modern concept of biological evolution arose before the 19th century. Darwin and Wallace provided the first plausible explanation for a mechanism of evolution by which species change through time and are connected in an ancestor-descendant relationship. Indeed, Darwin described many mechanisms of evolution including natural selection, sexual selection, kin selection, group selection, and inheritance of acquired characteristics. Whether any of these mechanisms act on a given population is a testable hypothesis, and many fields of evolutionary biology seek to assess their relative impacts. Indeed, the hypothesis that evolution is taking place at all in any given case is a testable, and therefore scientific, proposition. Darwin proposed many tests of evolutionary mechanisms and left us a legacy of testable hypotheses. The Darwinian view of biological evolution enjoys broad support from studies including the observed hierarchy of life (systematics), fossils (paleontology), developmental patterns (embryology), and most recently, genetics. However, support provided by embryology in particular has been called into question lately.
Support from embryology has been challenged throughout history in various ways by early critics of evolution including creationists, who now speak under the banner of "Intelligent Design." Behe (1998) and Wells (1999) claimed that embryological support for Darwinian evolution is based on the drawings of 19th century embryologist Ernst Haeckel (1834-1919). They reiterate Richardson et al.'s (1997) work indicating what has been known since at least 1894 (Sedgwick): Haeckel's embryological drawings are inaccurate and perhaps fraudulent. Accordingly, Behe (1998) claims that because of this, " ... the problem of development within evolution remains unsolved." Later, Wells (2000, Chapter 5) claimed that the embryological support for Darwinian evolution is based on the work of another 19th century embryologist, Karl von Baer (1792-1876). Wells says that as von Baer was not a proponent of evolution, so Darwin's use of his embryological contributions in support of evolutionary theory is in fact misuse. Later in the same chapter, Wells (2000) claims that von Baer's laws -- which describe continual change and specialization during development -- cannot accommodate reality because they do not allow for the Haeckelian notion of a conserved stage midway during development. As we will show, von Baer's view, and not the subsequent view of a mid-embryological conserved stage, is overwhelmingly supported by available data. Notwithstanding this, Behe (1998) and Wells (2000) both claim that the alleged existence of a conserved stage midway through development shows, also, that Haeckel's biogenetic law, "ontogeny recapitulates phylogeny," is false.
Wells and Behe claim that Darwin's thesis, as set forth in On the Origin of Species, relied upon embryological conclusions drawn before and after the publication of Origin. Further, they imply that current support of Darwinism is based on either von Baer or Haeckel, and that if embryological support for evolution comes from von Baer, then it is misuse; if it comes from Haeckel, it is fraudulent. While these simplistic claims are easily dismissed by embryologists and systematists, it can seem quite confusing to other biology teachers. We present the background necessary for teachers and students to evaluate this conflict objectively.
Here, we describe the history of modern embryology, the roles von Baer and Haeckel played in that history, and how that history relates to the formulation of Darwin's theories. In particular, we show:
A companion presentation (available in both Microsoft PowerPoint and Apple Keynote formats) for use by teachers that parallels the arguments discussed here is available via free anonymous download at ftp://www.biosci.ohio-state.edu/pub/IBP/.
Prior to 1830, embryology was an irregular science. Two early concepts were preformationism and epigenesis, both of which seem obviously false today.
Preformationism, formalized by Charles Bonnet in 1745, argued that all adult structures exist in the gametes, and that they merely unfold, or unroll as the individual develops. This, in fact, is the original meaning of the word evolution. An example of preformationism is the concept of the homunculus : a perfectly formed miniature person inside a sperm, who only needs to grow to become adult. Historically, many objected to this view of embryology: In many organisms (e.g., frogs) immatures do not resemble adults. Second, if preformationism were true -- if there was no such thing as development, but simply unfolding -- then the full-grown mother-to-be, so she must have in her eggs her own daughters -- and in their eggs, their daughters -- and so on, like nested Russian dolls, to infinity. Thus, the questionable necessity that all generations of humanity were preformed in Eve's body. In addition, preformationism did not account for the mixture of parental traits observed in offspring.
Epigenesis, literally "outside origin," is the opposite of preformation. Although today the term generally implies nothing more than gradual development from formlessness, as originally envisioned (by Aristotle, and later by 19th century embryologists), the egg is formless and outside forces determine its destiny. The egg, like a ball of warm clay, has only potential for its fate. Epigenesis, in its historical sense, was easily rejected by simple observations that, for example, chicken eggs incubated by a goose still become a chicken. The outside "goose environment" did not make a developing egg become a goose. Most aspects of the egg's identity were determined endogenously.
By the early 1830s, the scientific approach toward the study of development had taken place largely due to Karl Ernst von Baer. After meticulous examinations, in 1828 von Baer presented his concept of differentiation in detail in Entwickelungsgeschichte der Thiere (Animal Development). Von Baer's critical claim was that all early animal embryos appear similar and tend to look less so as time passes, and they develop specialized structures, with modification usually coming at the end of development. Members of close taxonomic groups tend to resemble one another longer through the developmental period. Species that differ according to phylum will differ early in embryology. Those that are in the same phylum but different classes differ later in embryology. Species that are in the same family are differentiable later still, and so on. The Linnaean hierarchy captures this temporal order of development and remains the strongest, most robust empirical support for the theory of descent from a hierarchy of common ancestors.
During von Baer's time, two overlapping schools of morphologists prevailed: a German Naturphilosophie school and the French transcendental morphologists. These movements asserted that organisms are organized in a chain of higher and lower types, and that this chain appears in embryological recapitulation. Formally introduced by the Naturphilosophie school (but with vestiges in writings of Aristotle [Gould, 1977]), recapitulation holds that organisms pass through developmental sequences including fully formed, adult stages of lower organisms. According to this scheme, human embryos, high on the scale of nature, sequentially look like the adults of fish, frogs, reptiles, and birds, before they acquire human morphology. Here "higher" and "lower" have no evolutionary meaning, but instead referred to improvement. (Some, even today, might erroneously see "improvement" as a natural consequence of evolution.)
von Baer opposed recapitulation. In the second volume of Entwickelungsgeschichte der Thiere, von Baer set forth a broad argument against recapitulation in the fifth of his six commentaries on the chick embryo. Here von Baer presented many objections to recapitulationism; two of the most damning were that (1) embryological features of lower animals can appear in adult stages of higher animals, and (2) defining characters of higher groups often appear early in the developmental sequence. The first point, today understood as paedomorphosis, is strictly incongruous with the notion that higher animals are improvements of lower animals. A good example of the second point is the amnion: Amniotes are defined by their possession of a derived trait, the amniotic sac of the egg. But, of course, this trait is found at the very beginning of the developmental sequence. In order for recapitulation to be correct, all lower animals should have an amnion as well. von Baer summarized his thoughts on development in his famous laws of development (as cited in Gould, 1977, 56):
von Baer's laws, including their explicit rejection of recapitulation, formed the basis for Darwin's discussion of embryology in Origin.
Darwin was heavily influenced by von Baer in forming his theory of natural selection (see review in Gould, 1977). Some choice quotes readily reveal this. Darwin refers to von Baer: "Hardly any point gave me so much satisfaction when I was at work on the Origin as the explanation of the wide difference in many classes between embryo and the adult animal, and of the close resemblance of the embryos within the same class" (cited in Gould, 1977, 1971). He wrote in 1860 in a letter: "Embryology is to me by far the strongest single class of facts in favor of change of forms" (cited in Gould, 1977, 1970).
von Baer did not accept evolution. Like most of his contemporaries, he gathered data from his observations and drew few inferences. It has been claimed (Wells, 2000) that Darwin's reliance on von Baer is invalid because von Baer did not accept evolution. But scientific revolutions necessarily interpret old data in a new light, and scientists who were leaders before the revolution are not so much converted as simply pushed aside (Kuhn, 1986). To say that any scientist preceding Darwin was not Darwinian is to state the obvious. For example, Einstein's theory of relativity is not compromised by the fact that Newton was not a relativist, even in light of Einstein's reliance on Newton's data. Indeed, the principal mechanism of scientific advance is simply new interpretation, through research, of standing data or concepts, and Darwin's use of von Baer's data is an excellent example. Similarly, Ernst Haeckel reinterpreted recapitulationism in light of Darwin's theory of evolution, but the fact that Darwin rejected recapitulationism does not invalidate Haeckel. Other data were required for that.
In 1874, 15 years after Darwin's publication of Origin, Ernst Haeckel published the most comprehensive treatment of vertebrate morphology to date, necessarily considering the theory of natural selection. Haeckel sought to fuse recapitulation with the relatively new theory of evolution. In doing this, Haeckel thought he found the mechanism of evolution: "Phylogenesis is the mechanical cause of ontogenesis" (Haeckel, 1874, 5; cited in Gould, 1977, 1978). Haeckel continued, "The connection between them is not of an external or superficial, but of a profound, intrinsic, and causal nature" (Haeckel, 1874, 6; as cited in Gould, 1977, 1978). To Haeckel, phylogeny was the engine of ontogeny, and thus ontogeny recapitulated phylogeny, his Biogenetic Law.
Haeckel promoted recapitulationism as intrinsically tied to evolution. In his 1874 treatise he proposed that all embryological innovations are added to the developmental sequence terminally. More ancient species should have a simple development and modifications to the development should occur at the end. Terminal addition causes more derived species to have more advanced, complicated, and extended development.
In order for this sequence of events to remain in the proper order, there could be no shuffling of the sequence. Accordingly, Haeckel believed the embryo was immune to selection. If this were true, then closely related animals would pass through an identical, identifiable stage of development, the phylotypic stage, which corresponded to the strict recapitulation of their most recent common ancestor. Haeckel pointed to the existence of this stage in his suspect drawings of vertebrate embryos.
Haeckel was wrong about his Biogenetic Law. This was not news, however, as even von Baer himself had shown 50 years earlier that recapitulationism is bankrupt when compared to actual data. The accuracy of Haeckel's illustrations has been challenged by numerous authors across more than a century, including Sedgwick (1894), Lillie (1919), de Beer (1951), Gould (1977), and most recently Richardson et al. (1997). Claims that Haeckel's errors are newly discovered, or have unrecognized significance (Wells 1999, 2000) are at best poor scholarship.
It should be clear from chronology that Darwin relied on von Baer but could not have relied on Haeckel. Nonetheless, Wells (1999, p. 349) states, "Of course, it would be illogical to conclude that Haeckel's distortions invalidate Darwin's theory. Although Darwin considered the embryological evidence 'second to none in importance' (Darwin, 1859, p. 346), he did not base his theory on that evidence alone" (emphasis added). Indeed, Darwin could not base his theory on any of Haeckel's views because they would not be published for another 15 years. Darwin considered embryology, not Haeckel's drawings, second to none in importance.
Richardson et al. (1997) found considerable variation among vertebrate embryos and concluded (p. 105):
These modifications of embryonic development are difficult to reconcile with the idea that most or all vertebrate clades pass through an embryonic stage that is highly resistant to evolutionary change. This idea is implicit in Haeckel's drawings, which have been used to substantiate two quite distinct claims. First, that differences between species typically become more apparent at late stages. Second, that vertebrate embryos are virtually identical at earlier stages. This first claim is clearly true. Our survey, however, does not support the second claim, and instead reveals considerable variability -- and evolutionary labiality -- of the tailbud stage, the purported phylotypic stage of vertebrates.
Haeckel argued that all vertebrate embryos pass through a similar stage of development, sometime midway in embryology. This notion is in contrast to the scheme of von Baer, which suggests continual divergence in form of various taxa during development. Clearly, Haeckel's conserved stage in development is mutually exclusive with von Baer's notion of continual development. Richardson et al.'s (1997) data (rejecting Haeckel's drawings and his concept of a conserved stage) strongly support von Baer's view of development, upon which Darwin relied.
Behe (1998) used Richardson et al., (1997) and references therein to argue " ... that for a century, Darwinism easily embraced a false description of a fundamental process and that the problem of development within evolution remains unsolved" and "the earliest stages of development are actually quite different across vertebrate species, and become increasingly similar toward the phylotypic stage. The hourglass pattern of development is a conundrum that is not predicted by Darwinism." Indeed, the hourglass pattern referred to is not predicted by Darwinism, but by Haeckel's false notion of the phylotypic stage (Richardson et al., 1997). And in any event, as discussed above, the validity of Haeckel's drawings, his concept of a phylotypic stage or hourglass model of development does not impact Darwin's hypothesis as these concepts were introduced 15 years after the publication of Origin. Instead, Darwin did rely on von Baer, who embraced neither the phylotypic stage nor the hourglass model.
Wells (1999) argued that Darwin's theory is partly invalidated because it relies on Haeckel. Interestingly, Wells (2000) subsequently claimed that Darwin's reliance on von Baer was invalid because von Baer's ideas are inconsistent with the Haeckelian notion of a conserved stage. Although Wells (2000, p. 98) acknowledges Richardson's (1999, p. 609) statement that "the phylotypic stage is a misleading concept that needs to be reassessed," he insists that "Nevertheless, no one doubts that vertebrate embryos start out looking very different, converge in appearance midway through development (though not at the same time), then become increasingly more different as they continue toward adulthood;" this in spite of Richardson et al.'s (1997) title: "There is no highly conserved embryonic stage in the vertebrates." The fundamental finding of Richardson et al. (1997) is that the phylotypic stage, and its corollary hourglass model of development, is unsupported. While neither the phylotypic stage nor the hourglass model form any basis of Darwin's description of the theory of natural selection, these are fundamental beliefs of Behe, Wells, and their students in their ongoing Intelligent Design Creationism attack on Darwinian biology.
Darwin did not rely on Haeckel, but rather on von Baer. von Baer's stance against "evolution" is irrelevant. Behe (1998) and Wells (1999, 2000) are deeply confused or intentionally confusing regarding the history and significance of this well-known field, an area they claim has special meaning in their political movement.
Behe, M.J. (1998). Embryology and evolution. Science, 281, 347-351.
Darwin, C. (1859). On the Origin of Species by Means of Natural Selection. London: John Murray.
de Beer, G. R. (1951). Embryos and Ancestors. Oxford: Clarendon Press.
Gould, S. J. (1977). Ontogeny and Phylogeny. Cambridge, London: Belknap Press.
Haeckel, E. (1874). Anthropogenie: Keimes- und Stammes-Geschichte des Menschen. Leipzig: Engelmann.
Kuhn, T.S. (1986). The Structure of Scientific Revolutions. New York: New American Library and University of Chicago Press.
Lillie, F. R. (1919). The Development of the Chick. New York: Holt.
Richardson, M. K., Hanken, J., Gooneratne, M.L., Pieau, C., Raynaud, A., Selwood, L. & Wright, G.M. (1997). There is no highly conserved embryonic stage in the vertebrates: implications for current theories of evolution and development. Anatomy and Embryology, 196, 91-106.
Richardson, M.K. (1999). Vertebrate evolution: the developmental origins of adult variation. BioEssays 21, 7.
Sedgwick, A. (1894). On the law of development commonly known as von Baer's law; and on the significance of ancestral rudiments in embryonic development. Quarterly Journal of Microscopic Science, 36, 35-52.
von Baer, K.E. (1828). Entwicklungsgeschichte der Theire: Beobachtung und Reflexion. Königsberg: Bornträger.
Wells, J. (1999). Haeckel's embryos & evolution: Setting the record straight. The American Biology Teacher, 61(5), 345-349.
Wells, J. (2000). Icons of Evolution. Washington, DC: Regnery Publishing, Inc.
Originally published (with figures) in The American Biology Teacher, Volume 67, No. 5, May 2005, pp. 276-282. http://darwin.bc.asu.edu/pub/pickett.pdf.