Revival of Extinct Species

Lonesome george died. Well, ok, that's only one extinct subspecies. What's more, it might not be gone forever.

Apparently, one animal has been un-extincted, though only briefly. The Pyrenean Ibex, which first died out in 2000, was subsequently cloned in 2009, though the clone lived for only seven minutes. Still, fairly promising. One factor in favor of reviving Lonesome George is the fact that only the particular subspecies was extinct. Subspecies can capable of interbreeding, that is, they are not differentiated by reproduction. In other words, other subspecies still exist, that are capable of surrogancy for cloning.

While that's fine and good, the thought of reviving extinct species made me wonder if we could indeed resurrect creatures even more far gone, where no direct genetic material, or no subspecies of the like, exist.

Typically, we might imagine evolution to be a one-way process, where species change with time into forms that are more adapted to the existing conditions. Such changes are brought about by recombination and mutation, the reshuffling and flipping of genetic material into new forms. Conceivably, the process is irreversible, or at least difficult to reverse. It is hard to imagine a modern creature evolving assuming ancient, extinct forms. 

Yet, it seems possible to revive extinct species, or some approximation of, through 'devolution' of a modern descendant. First, imagine if environmental conditions were reset to the time of the extinct species. If so, then the optimal adaptation is no longer of the modern evolved creature, but rather of the extinct form! Natural selection therefore favors the evolution of the modern creature towards the ancient, extinct predecessor. Just like how the final products of chemical reactions can vary with temperature and pressure, the direction that evolution takes can be redirected by varying the conditions under which evolution occurs.

Given a supervised hand, the revival of extinct species from their modern descendants can be made even more viable. If the path of evolution is known, for example the amounts of environmental change, and the corresponding change in the species, it is possible to perform directed artificial selection. By seeking incremental regression along the known evolution history of the species, one can create a ratchet-effect to gradually skip backwards.