"It has been a mainstay of evolutionary theory since the 1970s. Natural selection acts purely on the level of the individual and any cooperation observed between organisms merely hides a selfish genetic motive. There have been two pioneering theories to explain cooperation in the natural world given this framework: the first was William Hamilton's (1964) theory of kin selection and the second was Robert Trivers' (1971) theory of reciprocal altruism. However, both of these scenarios break down where it comes to unicolonial ants. In a paper in the journal Trends in Ecology and Evolution (subscription required) Heikki Helantera, of the University of Sussex, and colleagues at Rice University have investigated how previous theories to explain cooperation don't apply for these unique supercolonies." - james reilly from Bookmarklet

But aren't they an individual? Just like we are super-organisms. Just because there is thin air between something, does it mean its two objects? - Alexander Kruel

".. 'The extreme cooperation of unicolonial ants has been suggested to be an example of selection occurring on levels higher than the individual, such as the superorganism, group or even population.' ..Group selection is the idea that, under certain circumstances, genes will be selected for because they benefit the overall success of the group rather then simply the individual. While it is usually assumed that these populations will have a high level of relatedness.. the authors suggest a scenario in which group selection could apply even among unrelated group members." - james reilly

Often ant colonies differ from most multicellular organisms - since the workers and the queen are not 100% related. - Tim Tyler