THE BILATERIANS ARE split into two clades, the protostomes and the deuterostomes. These are differentiated based upon their embryonic development. While there are several differences (and variation within each clade!), the key difference is the fate of the blastopore. I was once presented with the question, “In protostomes the mouth develops from the blastopore, but in deuterostomes the anus develops from the blastopore, so how can you evolve a deuterostome from a protostome?” The protostomes include the insects and annelid worms, how can you reverse the development of those to get to deuterostomes like frogs and starfish?

The short answer is, you don’t! Let me count the ways:

  1. You might not have to in the first place–deuterostomes and protostomes might have independent origins, or deuterostomy might be the ancestral condition.
  2. The ancestral bilaterian would not have been as complicated as extant organisms, so messing around with its development would have been easier.
  3. This depiction of gastrulation is an oversimplification. Some protostomes produce both the mouth and the anus from the blastopore. If this is the primitive condition, deuterostomes just have to make a second mouth and close the first.

We often forget that evolution has proceeded greatly since the divergence of the protostomes and deuterostomes 550 million years ago. By our standards a beetle is simple. But the first beetles didn’t show up until 265 million years ago. We cannot determine how the deuterostomes evolved by starting with modern protostomes. We have to start with the ancestor of the deuterostomes and the protostomes, often called the urbilaterian ancestor. Originally the urbilaterian was thought to be similar to the flatworms, but they were discovered to actually come from more complex ancestors and have secondarily lost many derived features. Currently we are not sure what the urbilaterian looked like. We do know some things from studying the protostomes, the deuterostomes, and the cnidaria. We know that it probably was not segmented, had a tube-shaped gut, had a central nervous system, and had a variety of body-patterning genes which probably were a prerequisite for bilateriality and later morphological complexity.1 Most likely this ancestor lived in the Precambrian, and a possible candidate may be found in future among the weird and wonderful Ediacaran biota.2

There are three possibilities for the protostome-deuterostome split. The first is independent origin of both from an ancestral bilaterian, the second is evolution of the deuterostomes from a protostome-like ancestor, and the last is evolution of the protostomes from a deuterostome-like ancestor. This scheme3 shows these various routes, with the exception of ancestral deuterostomy (this diagram also shows the enterocoel theory, which is now considered unlikely since it requires a 90° axis shift and would produce gene expression patterns that are not observed).

Scheme Protostome to Deuterostome

An independent origin of the deuterostomes and protostomes is possible, but the most likely route is an initial protostome-like bilaterian leading to protostomes on one branch and deuterostomes on the other. This divergence would involve closure of the mouth inherited from the protostome-like bilaterian and opening of a new mouth on the opposite side of the body. Then a dorso-ventral inversion would produce the modern deuterostome bodyplan. This is considered the most likely route because of homologous gene expression patterns along the anteroposterior axis suggesting a common origin, and the swapped role of Bmp and Chordin in protostomes and deuterostomes.3, 4 Protostome and deuterostome bmp and chordin genes are homologous, but bmp is expressed dorsally in protostomes and ventrally in deuterostomes, while chordin is expressed ventrally in protostomes and dorsally in deuterostomes. This dorso-ventral inversion has recently been supported by the discovery of a corresponding left-right swap.5

Some have thought that deuterostomy might be the primitive condition because several protostomes (chaetognaths, bryozoans, and phoronids) possess deuterostome-like gastrulation which could be inherited from a deuterostomal ancestor. However, the bulk of the evidence suggests protostome-like gastrulation is ancestral. Since the slit-like blastopore proposed for the protostome-deuterostome ancestor is based on the gastrulation of some annelids and arthropods, this may be a primitive characteristic of protostomes inherited from the bilaterian ancestor, and leading to the evolution of deuterostome-like gastrulation in these unusual protostomes.6

Embryogenesis is a complicated and variable process, but recent developments in genetic expression in embryos have cast light on a divergence of the Bilateria occurring over 550 million years ago.


  1. Erwin, D. H.; Davidson, E. H. “The last common bilaterian ancestor.Development 2002, 129, 3021-3032.
  2. Budd, G. E.; Jensen, S. “A critical reappraisal of the fossil record of the bilaterian phyla.” Biological Reviews 2000, 75, 253-295. DOI:10.1017/S000632310000548X
  3. Holland, L. Z. “Body-plan evolution in the Bilateria: early antero-posterior patterning and the deuterostome-protostome dichotomy.” Current Opinion in Genetics & Development 2000, 10, 434-442. DOI:10.1016/S0959-437X(00)00109-X
  4. Gerhart, J. “The Deuterostome Ancestor.” Journal of Cellular Physiology 2006, 209, 677-685. DOI:10.1002/jcp.20803
  5. Hibino, T.; Nishino, A.; Amemiyah, S. “Phylogenetic correspondence of the body axes in bilaterians is revealed by the right-sided expression of Pitx genes in echinoderm larvae.” Development, Growth, and Differentiation 2006, 48, 587-595. DOI:10.1111/j.1440-169x.2006.00892.x
  6. Arendt, D.; Nubler-Jung, K. “Dorsal or ventral: similarities in fate maps and gastrulation patterns in annelids, arthropods and chordates.” Mechanisms of Development 1997, 61, 7-21. DOI:10.1016/S0925-4773(96)00620-X