It seems easy to overvalue these results. Speed dating call backs are cheap and shallow, so while the differences are interesting, I don't think they generalize well.
First, they should have had a scenario where both groups rotate, or alternate moving. It's not unthinkable that the mere act of getting up and moving around changes how one feels vs. sitting in the same chair for an hour or so. This would explain some of the results without any recourse to the theories proposed by the authors; interestingly, they do not make a serious effort at providing alternative explanations (or even explaining why, say, male interest fell when men sat - they just kind of assume it's naturally approaching the female average).
It's also significant that participants were all undergrads at a good school; this may make men more comfortable with being approached and women more comfortable with approaching, or it may just make them more conscious of this fact. It may be representative, it may even understate the effect, or it may not be representative at all.
Most significantly, speed dating really doesn't approximate reality that well, even if it is convenient for experiments. Who approaches whom can (and, in my experience, does) send important signals about personality and desirability, and these carry on much further than whether someone is willing to exchange contact info. So it's unclear how useful this data is for anyone who doesn't organize speed dating events, at least without further research.
I recently found an article in the New York Times that talks about a speed dating study that is going to be published in an upcoming issue of the journal Psychological Science. Given the usual state of science journalism, the fact that the article includes links that let me find a press release about the upcoming paper and a 20-page PDF file containing the paper itself was very helpful.
According to most studies and in accordance with popular stereotypes, men are normally less selective than women when it comes to evaluating potential romantic partners - in general, it appears that men are more likely to want to date any given woman than women are to want to date any given man. In a typical speed dating experiment, men and women rate potential partners as either a "yes" or a "no" depending on whether or not they want to see that person again. Men almost always rate a larger percentage of women as a "yes" than women do men, and, according to this paper, this is a fairly robust finding that generalizes over many different contexts. The usual explanation of this phenomena is based on evolutionary psychology: a female has a lot more to lose from a bad mate choice than a male does. If there were a biological, genetic basis for this tendency, it should be difficult to come up with an experimental setup in which women are less selective and men are more selective.
However, that's not the case at all. This study demonstrates that a small, seemingly trivial change in the speed dating ritual results in a (partial) reversal of the normal results. You see, in practically every speed dating setup, when it is time to interact with a new partner, men physically leave their seat and move to the table where the next woman is sitting, while the women remain seated and wait for the men to approach them. The authors of this study had the men remain still and had the women change seats, and found that this was all it took to wipe away the usual pattern: when the women were required to physically approach while the men remained still, the women became less selective then the men, reporting greater romantic interest and "yes"ing partners at a higher rate. "Rotaters" also reported greater self-confidence than "sitters", regardless of gender.
I suggest that you go read the paper, or at least the press release, yourself; my summary doesn't really do it justice, and I'm leaving the implications for the evolutionary psychology-based analysis of gender as an exercise for the reader.
EDIT: Having had some more time to look over the study, I think I should point out that it wasn't a complete reversal of the usual gender behavior: female rotators were only moderately less selective than male sitters, while male rotators were significantly less selective than female sitters. (Sitters of both genders were equally selective.)