Ferrofluid between glass plates: what’s going on? There’s a phenomenon I ran across recently and I’m trying to understand what’s going on with it. When you have a thin layer of ferrofluid between two glass plates and move a magnet closer and farther away from it, you get a really interesting transition between two states. When the magnet’s at a distance, the fluid breaks up into large droplets like this:

…but when the magnet’s closer, the droplets merge together into thin tendrils, a lot like the kind you get out of a reaction-diffusion process.

My question is: what forces are at play here? I assume the blobs/tendrils are held together by surface tension, but they seem to be repelling each other—are they acting as individual magnets? If so, what causes them to stay together at all instead of the intra-blob repulsion breaking them up?
 A: I think that the behavior can be understood by just considering the energies of the various configurations. In the initial state you have relatively large blobs because the ferrofluid is just mainly trying to minimize surface tension energy. But when you subject the sandwiched ferrofluid to a strong magnetic field, then a large amount of magnetization is induced in the ferrofluid since ferrofluids, like all ferromagnetic materials, have a very high magnetic permeability. So then we have the situation shown on the left side of the diagram below.

The problem is that this is not a very energetically favorable configuration because all of those "N" poles are trying to repel other nearby "N" poles, and all those "S" poles are trying to repel other nearby "N" poles. Much more energetically favorable for the fluid to break up into tendrils as shown on the right side of the diagram.
Notice that the same sort of thing tends to happen with solid ferromagnetic materials. The lowest energy state of a solid piece of ferromagnetic material is for the magnetic domains to arrange themselves into randomly oriented small domains. This arrangement minimizes the magnetic energy of the material. It's highly energetically unfavorable for a solid piece of ferromagnetic material to all magnetize as one big magnetic domain oriented in a single direction. The reason that such highly magnetized ferromagnets can exist is that the material is heavily doped with pinning sites which prevent the big magnetic domain(s) from breaking up and rearranging themselves into smaller, randomly oriented magnetic domains. 
A: the large droplets/blobby appearance is caused by the fact that the fluid probably does not "wet" the glass surfaces very well. In this case, the blobs of fluid act under the influence of their surface tension, which tends to pull them into round shapes and, if given the opportunity, to merge into larger circular blobs. 
In addition, the reason that some of those blobs look like little circles and the others resemble the squashed pancreas of a jungle lemur is that the closer together you press the glass sheets, the more you get the squashed-pancreas look, and the farther apart the glass sheets are, the more you get the circular blobs. 
Being a ferrofluid, that liquid is very strongly acted upon by your external magnet and despite being confined in a space where surface tension effects are dominant, the fluid tries very hard to trace out the directions of the field lines produced by the magnet. what you get is a tug-of-war between the magnetic and the surface tension forces, which looks like a lot of fun to fool around with!
