There are quite a few examples of organisms we consider "species" that are actually symbionts. Lichens, portuguese man-of-war, etc. Dr. Margulis, unfortunately, has gone way outboard by proclaiming that symbiosis is the
primary means of speciation. I don't think that contention is even remotely supported. Development of genetic incompatibility through geographic or behavioral isolation remains the key ingredient in speciation. (See for example, Orr 1995, "The Population Genetics of Speciation: The Evolution of Hybrid Incompatibilities", Genetics 139:1805-1813).
Where I think Margulis might have a better case is in early metazoan evolution - how did those first cells or multi-cellular critters evolve? In this case, symbiosis and serial endosymbiosis seems to be a pretty likely scenario.
And yeah, assuming termites are like ants in that respect, each species should have different gut bacterial symbionts that evolved along with their hosts.