Chiroptera writes:
Now why in the world would they publish the fact that they discarded the dates, for the entire world to read, if there was something fishy going on?
It may be otherwise but there's a pretty simple and obvious answer. Submit a paper with a "wrong" date and you don't get published. So even if one of the other dates is more strongly indicated, the safe bet is to publish the "correct" date, and then include the information on other dates for completeness.
Too bad it's for pay. I'd be curious to see how much camoflague, if any, was applied to the other dates. Might be they smuggled some truth past the gatekeepers.