A species is one of those things that is harder to define than it looks. While it's clear that (for example) a rat is a different species than a dog, the more closely related animals get, the harder it is to properly put them into species. Usual definitions involve the sharing of physical characteristics (although the physical characteristics shared between a Great Dane and a Shih Tzu can be harder to ascertain...) and the fundamental idea of breeding. If two species can interbreed to produce fertile offspring they are usually considered the same species. This still runs into problems; some animals can interbreed but just don't (for example if they live on opposite sides of a mountain range) some animals can interbreed and produce mostly asterile offspring with the occasional random fertile quirk, while some animals could potentially interbreed but it's rather hard to imagine (again, with the Great Dane and the Shih Tzu)
When it comes to the world of microorganisms, however, species becomes even more of a murky concept. For bacteria and archaea even the concept of interbreeding doesn't really exist. Bacteria don't require all that squishy 'sex' stuff to produce more bacteria, they just divide themselves in half and filch random DNA from their mates, nearby bacteria and the general environment. However at the same time it's still useful to be able to disinguish bacterial species; for example to recognise that S. aureus is a very different creature from an E. coli. All bacteria are not the same; but if they can share genes so easily, why are there such clearly different species? Why are there any differences between them at all?
Work done on archaea, the sadly ignored cousins of bacteria, is starting to give more of an idea about how a 'species' in microorganisms form. Studies done on the thermoacidophilic (high-temperature and low-pH loving) archaeon Sulfolobus islandicus gathered from a hot spring in Russia shows some mechanisms that, in these strains at least, leads to separate archaea species forming rather than one homologous population.
The paper highlights two main theories as to how species form in microorganisms. 'Species' in this instance is defined as lineages with discrete clusters of sequence diversity. Theory one is the 'niche' theory, the idea that selection is ecological, and "ecotypes" (i.e almost-species) are kept because they're specialising themselves to fit into a certain niche. If you're trying to be the best at living in a hot pond, genes to help you cope with ice-crystals aren't going to be much use.
The second theory is that there are genetic barriers that cause some clusters of micro-organisms to become genetically distinct and species-like. If the mutations within your own genetic material are more likely to be passed on than genes picked up from nearby organisms then it genetic clusters of species will form. This is all to do with the rates of mutation and recombination, and is explained much better (with many more paragraphs) in the reference below for those who are interested. It can also involve things like restriction enzymes, which chop up DNA. If one archaea line contains enzymes that break down the DNA of another line they aren't going to be successfully sharing DNA any time soon.
It could also be a bit of a mix between they two. If tiny amounts of DNA are being shared, while large changes are made to the chromosome due to niche-specialisation, then the species fits under theory one. If more DNA is being shared with fellow micro-organisms, and each one is not mutating much, then theory two is clearly more relevant. For different species, in different conditions, speciation may occur using different methods.
For the Sulfolobus islandicus however, there's now a lot more information on how species and clusters form. The researchers looked at two separate lineages within the hot springs, and found that although the archaea were capable of sharing DNA, they were none-the-less starting to become ecologically distinct and evolutionarily independent. Just to be clear, there was no physical barrier stopping them. They were all just swimming around in close proximity in the same warm springs. There weren't even biochemical barriers stopping them, the DNA was capable of surviving in both strains without being chewed up by enzymes or otherwise harmed.
So are these two separate strains 'species'? Not really, they still share DNA, but they seem to be on their way to becoming distinct species. Archaea, like bacteria, will always be capable of sharing DNA with each other. What is fascinating here is that sometimes, it appears, they just don't want to. It may be that species in micro-organisms form less because of barriers that physically prevent gene flow, and more because the organisms themselves simply don't want to share DNA anymore.
After all, if you've spent the last few hundred years specifically evolving to fit one precise little niche, how much help is DNA from someone who hasn't?
Ref 1:Cadillo-Quiroz H, Didelot X, Held NL, Herrera A, Darling A, Reno ML, Krause DJ, & Whitaker RJ (2012). Patterns of gene flow define species of thermophilic archaea. PLoS biology, 10 (2) PMID: 22363207
Credit for image 1.
Credit for image 2.
Credit for image 3.