Evolution is simply a change in gene frequencies in a gene pool
(and any associated phenotypic changes in the population over time
associated with this genotypic change). Natural selection is the
currently accepted model for explaining adaptive evolution in
genepools, i.e. that change in gene frequencies that leads to
the appearance of 'design' in living things. Evolution is something in
need of explanation, natural selection is that explanation.
The most important factor in promoting speciation is reproductive
isolation, either by physical barrier (allopatric), by the inviability
of certain heterozygotes compared to their homozygous parents
(parapatric), or by sexual (genital or behavioural) incompatibility
(sympatric).
Mutationism requires that mutations be consistently biased in the
direction of adaptation (by experiment, this is not the case);
saltationism requires that large jumps in genotype are occasionally
adaptive (by consideration of the n-space of all possible organisms, we
know that the vast, vast majority of all 'living' things would be very,
very dead, and most saltations would take us into the vast variety of
the unliving).
Evolution by natural selection requires variation between
individuals (as provided by mutation and recombination), heritability
of some of this variation (as provided by DNA), and a selection
pressure that kills less well adapted organisms (as provided by the
limited environment in which organisms capable of exponential
reproduction live).
Polymorphism in the banding patterns of snails might be kept high
by predation because predators will learn to hunt for snails by sight,
and uncommon banding patterns will be more likely to escape being
eaten. These differently banded snails will have greater reproductive
success, up until the point that the predators start noticing enough of
them to see if they are edible too. This is somewhat akin to the idea
of security by obscurity in computer programming.
A successful selfish replicator must have high fidelity (low
mutation), high fecundity (able to make many copies of itself rapidly)
and high longevity (not degrade rapidly).
Mutualism differs from parasitism because both partners in a
mutualistic relationship benefit from their association. Parasites
benefit at the expense of their host, and are very often much smaller
than their host, which differentiates them from predators, which are
generally larger the (or of comparable size to) their prey.
Genes in parasites transmitted in the germ-line will experience the
same selection pressures as the genes in their host's cells. This means
that what is in their host's interests will generally be the same as
what is in their own interests. This will lead to selection pressure
for genes that appear mutualistic.
'Nice guys [can] finish first' in an iterated prisoners' dilemma
because the iterated game allows the players to 'get to know each
other', allowing them to build up sufficient trust to play cooperate
rather than defect. Tit-for-tat is 'nice' in the sense that it does not
ever start a game with defection: such strategies do well against
themselves, and if this is coupled with immediate retaliation for
defection, they also fare well against nasty strategies.
A worker ant is only ½ related to her daughters, but ¾ related to
her sisters. This is a consequence of haplodiploidy, and means that a
worker is better off making new sisters than daughters to perpetuate
her genes. In bees, a similar argument would apply if the queen were
monogamous. She is not, and in bees, it is policing by (half-sister)
workers that prevents other workers from breeding.
Investigating the origin of life is made difficult by the fact that
it has only happened once to our knowledge. This means we are logically
allowed to chose very improbable events to account for it because we
have no other independent examples of life to compare ours to and no
idea of how many planets in a given million would produce something we
would call life (and the definition of life isn't too watertight
anyway). Furthermore, our knowledge of conditions on the early earth is
a little sketchy, and do not even know where we should place the
chemistry that produces life from 'lifeless' chemicals.