> So far nobody has commented on influences of salinity/ pH on sex ratios
There are probably several reasons for this, the main one being that no
one has really messed with them in that much detail. Further, most
people cannot sex them when the males aren't showing colour, hence all
the stories about how they change sex. Rob Wager thought he had had some
change sex, but when he tried to replicate it he couldn't. It is
possible that they will change sex, however, no one has come close to
demonstrating it yet.
Given the environments they live in I would postulate that pH/salinity
wouldn't have too much influence on them. They live in almost anything
as long as it is wet irrespective of water conditions. You usually find
them in abundance when they are present, thus, if certain conditions gave
you mostly one sex (especially males) you may see them wipe themselves
out in that waterbody, ie if 90% were males, the 10% of females would get
harrassed severely and wouldn't likely live very long, thus reproduction
would be low to nonexistant. Of course, that is an extreme example,
minor shifts in sex ratio wouldn't likely be a big deal.
A common strategy amongst fishes is for males to look like females, ie be
subdominant to avoid getting the crap beaten out of them by other males.
When the dominant male dies, one of the subdominant ones quickly takes its
place. This is what is most likely happening. Some fishes take this to
extremes such as north American sunfishes. Sneaker males look like
females, and they will enter the nest of a spawning pair and is ignored
by the male as he thinks the more the merrier. The sneaker male deposits
it's sperm on the eggs of the female and leaves. The dominant males then
goes ahead and raises his children! Sounds like a common strategy
amongst some mammals too......