The recent discovery of animals that appear to live entirely without oxygen has confirmed a scenario of convergent evolution in the development of hydrogenosomes, demonstrating with near certainty that mitochondria have evolved into hydrogenosomes multiple times. This was already pretty will demonstrated by the discovery of hydrogenosomes with mitochondrial DNA,   as well as the fact that they, and mitosomes (similar organelles that do not produce hydrogen) "all share one or more traits in common with mitochondria (Fig. 2), but no traits common to them all, apart from the double membrane and conserved mechanisms of protein import, have been identified so far."
These animals are members of the phylum Loricifera, which is (distantly) related to arthropods and other members of the general taxon Ecdysozoa. Members of this phylum tend to have very complex life cycles (for metazoans), with at least some species having a parthogenic larval stage intervening between stages of adult sexual reproductions.
This is an exciting discovery, both in terms of the potential discoveries in energy biochemistry and what it says regarding the overall evolution of the Eukaryotes.
Efforts to pin down the exact "evolutionary tree" of the early Eukaryotes have, more and more, shown a tangled relationship among various proteins (and their coding genes), implicating a large amount of lateral transfer. The role of mitochondria has changed during this process. At one time the various amitochondrial Eukaryotes were regarded as (possibly) descended from the ancestral premitochondrial Eukaryote. By now, however, it's pretty clear that most (probably all) of these lineages are descended form mitochondrial Eukaryotes.
To complicate the picture, a number of (probably distantly related) lineages possess mitochondria that are facultative anaerobes. Among these lineages are several animals such as parasitic helminths such as Fasciola hepatica and Ascaris suum. It seems plausible that the probable hydrogenosomes of these newly discovered Loricifera are descended from such facultatively anaerobic mitochondria. (Or more precisely, descended from the original mitochondria via such facultatively anaerobic mitochondria.)
It's interesting to speculate regarding the specific metabolic pathways these species use. One possibility is that these organelles, despite looking like hydrogenosomes, are actually using sulfate as an electron donor, producing hydrogen sulfide. Another is that they actually produce hydrogen, which is then used by other organisms for energy, to reduce sulfates to hydrogen sulfide. Certainly the former would provide more energy for a multicellular creature. OTOH it would also have (probably) required a longer evolutionary path to reach. The question I suppose, is whether a simple hydrogen-producing metabolism could have provided enough energy for such a creature. It certainly seems plausible that a sulfate (or sulfite) reducing metabolism could have evolved in a facultative anaerobe, followed by streamlining into an obligate anaerobe.
I'm certainly looking forward to the publication of further research on these animals.
T/H Nick Anthis
Danovaro, R., Dell'Anno, A., Pusceddu, A., Gambi, C., Heiner, I., & Kristensen, R. (2010). The first metazoa living in permanently anoxic conditions BMC Biology, 8 (1) DOI: 10.1186/1741-7007-8-30
1 The first metazoa living in permanently anoxic conditions Open Access (Preliminary PDF)
2 Anaerobic Metazoans: No longer an oxymoron Open Access (Preliminary PDF)
3 Anaerobic animals from an ancient, anoxic ecological niche Open Access (Preliminary PDF)
4 Organelles in Blastocystis that Blur the Distinction between Mitochondria and Hydrogenosomes Open Access
5 An Introduction to Loricifera, Cycliophora, and Micrognathozoa Open Access
6 Degenerate mitochondria Open Access
7 Eukaryotic evolution, changes and challenges
8 Mitochondria as we don’t know them