I think it is worth examining this issue before going any
The discussion of whether ecosystems succeed in maximising
energy flow through them in the long term has been
previously discussed under the rubric of the "prudent
A group of prudent predators refrains from consuming too
many prey - in order to preserve a healthy breeding
population, for use on future occasions.
This necessarily involves individual predators refraining
from destroying prey - in order to avoid depleting prey numbers
to the point where the next generation of predators suffers
If the predators succeed in refraining from over-
exploitation of their resources, they are described as
"prudent" - and making "efficient" use of their prey.
The benefit of eating one of the few remaining prey
individuals is immediate and accrues to the individual who
eats it. The cost of not being able to eat all that
organisms descendants is less immediate - and is borne by
It appears that many predators are not prudent. They selfishly
maximise short-term benefit to themselves - and drive the
ecosystem they are part of into a cycle of fluctuations which
resucts in a less stable population for all concerned.
Tragedy of the commons
The tragedy of the commons 
arises when a resource is available to
all, but an individual gains from his exploitation of it -
wheras the cost of having the resource exploited is paid by
The classic example is a pasture. Individuals benefit from
having more animals grase the pasture, but having the pasture
over-grazed causes everyone severe problems.
A common result is that everyone tries to reap maximum
benefit from the pasture by grabbing as much for themselves
as possible - but the result is that the pasture is soon
depleted - and the animals all starve.
It can be seen that the problem of the prudent predator is really
a type of "tragedy of the commons" problem - with the additional
complication that the resource is self-reproducing.
There are opportunities available to them which would help them
be more efficient:
Selection on higher levels
While there seems to have been some success in using models
of group selection to illustrate that the prudent predator
is a possibility (e.g. ),
group selection has several "traditional" problems that
seem likely to result in this solution not being widely
In particular, the formation of distinct groups in nature
depends of gene flow between the populations being very low.
In sexual populations, groups may arise and fall only rather
If differential group reproduction is insufficiently rapid
there is a danger that individual level selection
will destroy the variability between groups that group
selection needs to work.
Selection at higher levels has the best chance of working when it
can prevent variation in individuals for the trait in question
from occurring - by controlling how they are constructed.
It seems that it would have a hard time doing something like that
for complex, social traits - such as prudence.
I think populations will eventually succeed in coming to terms
with the "inefficiency" represented by these problems - but
I don't think it should be relied on to sort the problem out -
and think that the main solution lies elsewhere.
Collective problems sometimes require collective solutions -
and one way of dealing with these sorts of problem involves
monitoring of the behaviour of individuals by
representatives of the population - and punishing predators
who are not prudent.
Provided the cost of the monitoring is not too large - and
is sufficiently widespread - this strategy can be effective
at reducing the frequency of selfishness.
This sort of mechanism can be seen in action in some places
where "fishing quotas" have been imposed by authorities in
an attempt to manage populations of resources that can be
This type of solution has been described in some detail by Matt
Ridley - in his book, The Origins of Virtue: Human
Instincts and the Evolution of Cooperation .
One way to resolve problems caused by a system being made up of
components with different goals is to make the system more uniform,
eliminating the conditions that cause the conflict completely.
My example of doing this would be human mitochondria.
Mitochondria have different reproductive paths to the
nuclear DNA. This means, for example, that it is normally
in the interests of any mitochondria in male humans to
commit suicide - causing any parental investment to be
diverted into females.
It also means that selection between the (oftem thousands
of) mitochondria in human cells might favour traits which
are counter-productive for thi individual thay occur in
Engineering the mitochondria's genes out and putting them on
the nuclear chromosomes would probably solve a lot of health
problems - if it could be done.
This sort of thing can usually be done - at least in
principle - by substituting a collection of organisms with a
single, more harmonious super-organism.
In the future, such collective arrangements seem more likely
to be nipped in the bud. These days there is an alternative
to symbiosis - if you want to take advantage of an adaptation
located by another creature - you can steal their genes.