It depends exactly how strictly you define speciation. If you follow the 'weak' version of 'these two populations will not normally breed with each other and cr

It depends exactly how strictly you define speciation. If you follow the 'weak' version of 'these two populations will not normally breed with each other and create viable offspring' then we've done it a few times with flies [1]. We've created populations that will not normally breed with each other. If you pick a stronger version, where they -cannot- breed, even if you were to preform in vitro or something similar on them, then we haven't quite done that yet. [1] http://www.jstor.org/pss/2410209 http://www.jstor.org/pss/2410209 (overview of attempts to record speciation in a lab)