Nature has a way of screwing with the human desire for order and simplicity. One of the most egregious examples I’ve encountered is the contrast between what you learn about animal development and, well, reality.
Bilaterian animals have been traditionally divided into two great groups: protostomes and deuterostomes. If I whip out my trusty little animal phylogeny, deuterostomes are indicated right there in the middle, and protostomes are the sum of the ecdysozoans and lophotrochozoans. (Whoever came up with the latter name, urgh. I have to think twice before I say or type it, and I’ve been into them for years.)
The names “protostome” and “deuterostome” were nicked from ancient Greek and literally mean “first mouth” and “second mouth”. They refer to where the mouth comes from during the development of these animals. Animal embryos go through a process called gastrulation, during which the initially ball- or disc-like embryo folds in to form a pocket, the primitive gut or archenteron. The archenteron connects to the outside through a single hole, the blastopore. In protostomes, supposedly, the mouth comes straight from the blastopore, hence “first mouth”. In deuterostomes, the blastopore forms the anus while the mouth opens somewhere else. (Or that’s what they teach you at school anyway)
Here’s a really cool animated gif of gastrulation in sea urchins from Stanford University’s sea urchin embryology resource. Note how the mouth appears late in the process and joins up with the archenteron – sea urchins are good and proper deuterostomes! (The red dots are the cells that form the larval skeleton, in case you wondered)
And here are real sea urchin embryos before and during gastrulation (source):
(Sea urchins are neat.)
That sounds nice and simple and clear-cut, which should be a big warning sign that it’s too neat to be true. As far as I know, deuterostomes are relatively well-behaved in this respect, but protostomes… protostomes are horribly misnamed. Hejnol and Martindale (2009) compiled a table of what’s known about the various openings in different bilaterian phyla, and found protostomes to be all over the place. In protostome embryos, the blastopore may become the mouth, the anus, both or neither, and this often varies even within a phylum.
In Hejnol and Martindale’s book chapter, priapulids (a. k. a., and I’m not kidding, penis worms) are listed as “?blastopore closure”. Well, when Andreas Hejnol and friends looked more closely at priapulid embryos, that became “nope, definitely deuterostomy”. Not only does the priapulid blastopore become an anus, its surroundings also express genes associated with butthole formation, and the “mouth” genes are active on the opposite side of the embryo. (Below: Priapulus developmental stages, from Martin-Durán et al.  The forming mouth [mo] and anus [an] are marked by shiny green concentrations of actin protein.)
That’s not just another addition to an already long list of deuterostomous protostomes, the study argues. Priapulids are considered to be one of the more conservative phyla among the ecdysozoans. With nematodes for comparison that’s not saying much, but this study suggests that they are quite conservative in terms of gut developmental genetics. The authors also note that deuterostomy was likely the ancestral condition for both nematodes + nematomorphs and arthropods + water bears + velvet worms (I’m not sure how strong that inference is based on what they write about the above groups). That means it’s likely to have been the way the last common ancestor of ecdysozoans developed. Given that deuterostomes are deuterostomous, the ancestral ecdysozoan probably was, and arrow worms, a weird ?protostome phylum that’s probably neither ecdysozoan nor lophotrochozoan, also are, this seems to suggest that all bilaterians came from a deuterostomous ancestor. (Below: the front end of an arrow worm, just for the heck of it. Yvan Perez, Wikimedia Commons.)
And that speaks against one of the more popular theories on how bilaterians came about from a jellyfish-like ancestor. Cnidarians such as jellyfish and coral polyps have only one gut opening, which is derived from the blastopore. A popular (and, IMO, quite appealing) idea is that this opening elongated in the ancestors of bilaterians, and separate mouths and anuses came from the long slit-like blastopore closing in the middle. If the last common ancestor of bilaterians is shown to be deuterostomous, this proposition remains without evidence.
There’s always a catch, though. Remember what I (or rather, Hejnol and Martindale) said about variation within phyla? Well, priapulids today are not the most diverse phylum to put it mildly, but there are still sixteen known species in two extant classes. Martin-Durán et al. (2012) examined one. You know the obvious question: how do the others develop?
It occurs to me that with so few living species, priapulids are among the few phyla for which this question could be answered completely 😀
Hejnol A & Marindale MQ (2009) The mouth, the anus, and the blastopore—open questions about questionable openings. In: Telford MJ & Littlewood DTJ (eds.) Animal Evolution: Genomes, Fossils and Trees. Oxford University Press, pp. 33-40
Martin-Durán JM et al. (2012) Deuterostomic development in the protostome Priapulus caudatus. Current Biology in press, available online 25/10/12, doi: 10.1016/j.cub.2012.09.037