I only just caught back up with this as I was out most of last week.
To clear things up somewhat (I hope):
First off I want to note that I am speculating in relation to Neps. Most of the info I am going off is what I remember from what popped up on the web when I was having a conversation with James Cokendorfer (not sure on that spelling) a few years back and the majority of that was related to Cannibis because those people seem really happy to talk about it all over the place.
What I found was that the sex of Cannibis (and therefore I assumed dioecious plants) is determined early in growth by exposure to "stress" and once determined it would not naturally switch. So in the case of Cannibis, if the early germinated seedlings encounter any kind of "stress" like less than ideal water levels or light levels or heat then the majority of them will turn out male. The flip side being that germinated seedlings under ideal conditions will all be female. The causative agents responsible for this were concentration gradients between GA3 (like the Maize note above) and one other that I think was called cytochinin or something like that. Males would develope under low GA3 levels and females under high GA3 levels (compared to the cytochinin).
I was discussing this with James because we were talking about seeing if it would be possible to get an already committed plant to "switch" sex for the purpose of cross breeding for greater genetic viability with N. clipeata. The theory was that if there were more pure female clones than male clones that perhaps one or some of the females (or cutting from the females) could be induced to "switch" sex thereby providing a new pollen donor and a larger gene pool. I don't know if James ever tried to follow through with it though... Maybe I should drop him a line...