A dark galaxy might seem at first like an oxymoron : how can you have a galaxy without any stars in it ? Surely that would be like having a zoo without any animals or a hospital without any doctors !
Not necessarily. Probably the best way to distinguish galaxies from other systems is their dynamics - that is, the way in which their stars and gas are in motion. In gas-rich nebulae, motions are dominated by gravity, stellar winds and supernovae explosions. Conversely in globular clusters, which are made purely of stars, the motions can be entirely explained by gravity alone.
But in galaxies it's different. When we measure their speeds, we typically find that their stars and gas in their outermost regions are moving many times faster than we would expect based on their visible material. Something massive needs to be present to accelerate them to these velocities, and equally, something massive is needed to stop them flinging themselves apart. We see similar effects on other scales too, such as in whole clusters of galaxies.
This "something" is given the inglorious but mysterious placeholder name of dark matter. We know quite a lot about what it does - altering galaxy rotations, facilitating the formation of the much larger structures we see in the Universe, enabling the rapid growth of galaxies through mergers - but precious little about what it is. We (more or less) know that it exists, because we see its effects in so many different conditions in an astonishingly self-consistent way, but so far we've never seen any signs of it emitting radiation. Hence it's dark. Optically dark, at any rate.
Most galaxies seem to be dominated by dark matter by a factor of a few compared to their visible mass. In some cases this might be a factor of ten or even a hundred. A dark galaxy simply takes this to its logical extreme : a dark matter "halo" without any stars at all. This sounds pretty useless because we'd never be able to detect it directly... unless it also has gas. And that's where radio astronomy, my main research topic, comes in.
Radio telescopes like Arecibo can detect the emission from atomic hydrogen gas at 21 cm wavelengths. If we're very lucky, some dark galaxies might have just enough gas to be detectable in our radio surveys, but not so much that they're ever formed any stars. This isn't the only method we have to search for such objects (other people have tried looking for unexplained disturbances in galaxy structures; gravitational lensing can also be used), but it's the method I'm using. It has the advantage that there's an awful lot of other stuff we can learn about galaxies by studying their gas content, even if we don't find any dark galaxies.
But is there any reason to think dark galaxies might exist, or are we doing this for shits and giggles ? In fact they could potentially be very important indeed. Cosmological models have been fabulously successful at predicting the observed structure of the Universe on the largest scales, but have some miserable failures on the smallest scales. The prime example of these is the missing satellite problem, in which models predict far more dark matter halos than observed galaxies. Dark galaxies are now one of the leading ways to explain this, but very few good candidates have ever been found - and all of those are intensely controversial and tend to make a lot of people quite angry.
In the following pages I give brief, quick introductory summaries of the major candidates that my research has been involved with (not excluding unpublished findings). They contain links to much longer, more detailed blog posts, and if you want the full technical details of the research, see the publication section. For my thoughts on the philosophy of dark matter - whether we're just making up nonsense and actually it points to some glaring error in our basic theories - see the Writings section.