Firstly I'll clarify that you are talking about simultaneous hermaphrodites rather than sequential hermaphrodites (1st one sex, then the other e.g. the limpet Patella vulgata).
It is perhaps easiest to address the question by countering it and asking why dioecy (2 sex systems/2 gonochoric types e.g. male and female) is better? As you have pointed out there are obvious advantages to being a hermaphroditic species such as more chance of mating - more likely to provide an advantage at very low population densities where interactions are infrequent.
There are two key disadvantages of hermaphroditism which I will briefly cover but have been discussed in this paper and probably other costs.
The first is energy costs. Maintaining the capacity to produce male and female gametes will be more costly than maintaining one. This gives the hermaphrodite a fitness disadvantage because energy is rarely an infinite resource. Therefore at higher population densities, when mating opportunities are not rare, the gonochoric individuals will have a higher fitness because they have more energy. Monogamy is also analogous to rare encounters but true monogamy is rare (1 partner for life).
The second disadvantage of hermaphroditism is self fertilisation. This will cause an increase in homozygosity and lead to inbreeding depression (reduced fitness).
So you are right to some extent...
the hermaphrodite way of reproduction is more successful than the
"normal" way.
...but the conditions which give rise to an hermaphroditic advantage are restricting. Overall, the above costs, combined with the obvious complexity of evolving the ability to produce male and female gametes, the ability to both fertilise and be fertilised, pregnancy and birth, and mating systems, mean that it is often more beneficial to be a dioecious species. Thus dioecy evolves.
EDIT: Question Raised by @Single_Digit
I have been pondering this question for a while and I get what RG255
is saying. I'm just not sure I entirely buy it. Take earthworms, for
example. They are simultaneous true hermaphrodites (as far as I
understand). The anatomy doesn't have to be that complex*. They simply
have two genital openings (one for eggs and one for sperm) and they
line up in a "69" (excuse the vulgarity) position. This should, in
theory, minimize the inbreeding depression. However, it doesn't
eliminate the maintenance of two sets of reproductive systems. But
most organisms are not internally fertilized mammals with wildly
complicated systems of internalized embryonic care. Most species lay a
pile of eggs that a male squirts sperm on or squirt eggs while the
male squirt sperm and then they hope for the best.
I would think the advantages of simultaneous parenting (after all,
many MANY species' males don't provide much in the way of child care)
and its fitness advantage would vastly outweigh the burden of a second
set of reproductive organs.
With that said, I don't have a better explanation, but I find the
question a very interesting one. The linked article is pay-walled,
aside from the abstract, but I still disagree with some of its tenets.
To me true hermaphroditism should be very common (I realize it isn't)
in species that don't need two parents to raise offspring, but do
benefit from some (as in one parent's) parental care. I recognize that
it would do little to help species that merely dump gametes and leave
because specialization of one reproductive system would likely do the
job better and both genders equally contribute under that type of
system.
So, RG255 convince me! Clearly there are good reasons, since gender
(or asexual repr) is the norm, but I need more/better evidence.
- Yes I realize they would need separate internal anatomies for each type of gamete, but still...
My response:
You have presented one example of hermaphroditism and used that as
evidence that all species should be hermaphrodites. Earthworms are
small slow creatures living in soil, I don't imagine they have high
rates of encounter, and therefore low rates of encountering the
opposite sex, therefore hermaphroditism would be favoured as discussed
above.
Further, you say most species are external fertilisers (do you
have a reference for this?) and therefore it is not costly be a
hermaphrodite. I don't see your logic there, the cost is not
necessarily to do with the cost of bearing child, producing &
maintaining the gonads and gametes is also a costly process. I would
argue that this is extremely complex. This is not just on a
morphological level but also physiological: in non-hermaphroditic
species the sexes have very different, and often, conflicting gene
expression and hormone production patterns. Hermaphrodites would not
be able to optimize to the fulfilling both the male and female roles.
Finally, you pointed out that the worms do not inbreed. Inbreeding
avoidance does not have to be the cause of the hermaphroditism
persisting, if the environment/other factors favour hermaphroditism. I
never said that both were simultaneously necessary.
I hope this
clarifies it for you, if not please expand as to why, I am on here
because I want to help people understand biology properly!
Further response from @single_digit:
Well fair enough about my external fertilizers comment. I don't have a
reference, but I was thinking all multicellular life and I'd have to
imagine that when you factor in plants, that external fertilization is
relatively the norm (as is hermaphrodism (dioecy) for the plants). As
to earthworms, I disagree about your description of them. Their
densities are actually pretty high, so I'd wager they encounter each
other frequently, so I'm not sure where that leaves them in terms of
pressures for hermaphroditism. Your point on the physiologic/hormonal
issues of maintaining the systems is one I haven't previously
considered. I honestly don't have any clue as to how daunting (or
simple) that is, but I'd imagine that the sophistication of the
systems would play a pretty key role. Makes me wonder how much this
has been researched in true hermaprhodites. I suppose the main thing I
keep coming back to is the overwhelming disadvantage gender has in
terms of potential to create offspring. Males in many (most?) species
essentially act as little more than sperm donors, thus half the
individuals have effectively zero fitness. That just seems like an
overwhelming advantage for hermaphrodites.
My Response
Why do you consider half of the individuals to have zero fitness?
Fitness is widely accepted as the number of offspring a parent
produces because this is directly related to number of copies of their
genes passed to the next generation. Sperm donor type males achieve
increased fitness by mating as do females - with out the male they
would never be fertilized. The key disadvantage of dioecy is the
halved (assuming equal sex ratio) frequency of potential encounters
that could lead to mating. The general disadvantages of
hermaphroditism are inbreeding depression and high cost & complexity.
Single_digit:
Zero fitness isn't exactly correct, but if we look at parental care as
conveying a survival advantage for K selected species, and huge
numbers of offspring conveying an advantage for r-selected species
(obviously the type of env affects this) does a deadbeat dad really
optimize for either of these? Passing on genes is fine, but if
offspring survival is low, does it matter? Does it simply boil down to
the maintenance of two repr systems plus decreased fitness from
inbreeding vs the increased reproductive success from extra child
care? Or is there more?
My response:
r/K selection theory has generally been disregarded in the
evolutionary biology community due to the substantial evidence against
it so it is unhelpful to think of selection in this way. As long as
the 2 sexes strategy is more successful at passing on genes than a
hermaphroditic strategy it will (should) prevail. Dioecy will be more
successful if the hermaphroditism introduces to much cost through
production and maintenance of sexual organs/gamete and inbreeding
whilst not attaining substantial gains from higher potential mating
frequency.
@Single_Digit
Interesting about r/K selection. I hadn't heard that. Do you have any
links? I'd be curious to learn more there. I incorrectly earlier made
a comment about dioecy where I meant monecy. But this seems to beg the
question, why is monecy/hermaphroditism so much more prevalent in
plants? Obviously there are different survival pressures, but I'd
think the same basic principles would apply as in animals, but the
condition seems to be far more common than in animals.
My Response
I seem to remeber there being a reason in plants, don't
have time to look it up right now. The work about r/K selection was
Reznick/Stearns/Charlesworth. Reznicks is the most recent and more
overview type paper - best place to start:
http://www2.hawaii.edu/~taylor/z652/Reznicketal.pdf