Re: is evolution truely random?

Evolution occurs through natural selection acting on individuals to modify
the chance that their DNA is passed on to successive generations, all
things being equal. One example of natural selection would be when males or
females judge aspects of their mates to be “attractive” and this influences
the number of offspring an individual produces. Research with birds and
people has shown that symmetry of features is a very desirable trait in a
potential mate. Another example can be found in the work of Peter and
Rosemary Grant of Princeton University. They measured a number of physical
characteristics of species of finches in the Galapagos over many years and
many generations. By coincidence an El Nino event occurred in the midst of
their work. This caused the available food for the finches to shift from
seeds that were easily cracked open to a variety that was tougher to open.
The Grants found that very soon after the food supply changed the beaks of
the finches changed to be more robust, which would allow these birds to
feed on the tough seeds. Variation in beak robustness (a function of length
with respect to depth) was always present in the finches. It was when the
food source changed that the mean “robustness” of beaks changed. Individual
birds with beaks that could not open the tougher seeds would starve before
they could reproduce. This is an example of selection acting on a physical
characteristic and thereby changing this characteristic in a population.
Even though the variation measured was anatomical the basis for this
characteristic had a strong genetic component.

This last example illustrates an important point, that individual genes
exist in different varieties among individuals in a population. These
varieties of a specific gene are known as alleles. If an allele has a
detrimental effect on an individual’s ability to produce offspring (as in
the case of Galapagos finches with slender bills) then natural selection
will, over time, reduce the frequency of that allele in a population.
However, this may not be strictly true under certain conditions, such as
small population sizes, where an individual with a detrimental allele may
produce offspring simply because others have to mate with him or her to
produce any offspring at all. In general, and especially when one thinks of
genes at the molecular level, most alleles aren’t detrimental at all to the
production of offspring. This variation is known as neutral variation, and
the persistence of neutral variation is to a very large degree a random
process. In fact, the probability of neutral alleles persisting in a
population is directly proportional to the probability of mutation. 

This leads back to another aspect of your question and that is mutation.
Changes in DNA sequence from one generation to the next come about due to
mutation and humans will pick up on average one new mutation to their
genome each generation. The rate of mutation can be increased by
environmental factors such as exposure to radiation or certain chemicals
(teratogens or mutagens), but there is always a background rate of
mutation. As mentioned above, much of mutation is neutral and selection
will not affect these alleles' chance of being passed on to future
generations. Rarely, mutation will affect how a gene product performs its
task and this is when selection occurs. In humans selection often takes the
form of inheritable diseases such as Tay-Sachs or cystic fibrosis which
affect the patient’s ability to reproduce. However much these diseases
reduce one’s ability to have children the genes themselves don’t disappear
from the population, but the alleles that lead to the disease remain fairly
rare. There may even be a selective advantage for having one copy of a
disease allele (among the two copies of every gene that one has in his or
her genome), as has been postulated for the persistence of the allele
causing sickle-cell anemia in human populations subject to malaria. Indeed,
if the gene responsible for sickle-cell anemia was not present then a
viable embryo could not form as this gene codes for a component of
hemoglobin. Likewise with cystic fibrosis as this gene codes for an
ion-channel protein in cell membranes, and having one cystic fibrosis allele 
may reduce the effects of cholera. 

So, is evolution random? Sometimes it is quite random and other times less
so. It all depends on the genetic and environmental context of the trait of
interest.