>     is death a universal reality?  For the mystery of birth
>     and the mystery of death, evolution has no answer.

SEXUAL REPRODUCTION:
  I'm not sure _when_ in the history of life sexual reproduction
became common (no one is), but I can explain (far too briefly) a
couple of hypotheses as to _how_.  [See _The Masterpiece of Nature_,
by Graham Bell; _The Evolution of Sex_, by John Maynard Smith; and
_The Evolution of Sex_, by Rick Michod & Bruce Levin (yes, 2 books by
the same name); also, an issue of _Discover_ from last year had a
passable article on the topic].
  The question of how sex _started_ is different from the question of
how it is _maintained_ in the face of selection.  It _started_ merely
as a process of DNA transfer between individuals, possibly as a
mechanism of gaining a template for DNA repair (see the Michod & Levin
book) or as a result of a bit of ultra-selfish DNA (from a talk I saw
by Graham Bell).
  Sex is _maintained_ by one of two classes of mechanisms, or a
combination thereof:  genetic mechanisms, and ecological-genetic
mechanisms.  Genetic mechanisms are postulated because outbreeding
(and recombining with a dissimilar genome) can reduce the number of
deleterious mutations in the offspring - since recombination
"shuffles" bits of the two parental genomes, some of the offspring
have fewer deleterious mutations.  So, under sexual reproduction the
average fitness of the population is higher than under asexual
reproduction (WARNING:  Group selection argument - which is okay, but
I just wanted to point it out).
  According to ecological-genetic mechanisms, the advantage of sex
still comes from the shuffling of the genome, but the fitness effects
are due to ecological, rather than intrinsic genetic, factors.  For
example, if parasites infect their hosts by matching some genetically-
determined characteristic (i.e. if they can evade the immune system by
mimicking a host protein), then there's going to be constant
coevolution between the host and the parasite:
     The parasite population will evolve to infect the most common
     host type.  This, however, makes the _most common_ type into the
     _least fit_ type, because it's being infected most often.  So,
     that type decreases in frequency, with a corresponding increase
     in frequency by another type.  Of course, the parasite then
     evolves to infect the _new_ common type, and the cycle begins
     again.
Anyways, the point is that there is an advantage to the individual to
be able to produce rare offspring.  Sex allows this to occur.
  Note that there are many subtleties to, and variants of, these
hypotheses which I haven't even begun to touch on.  Despite the hints
dropped by the _Discover_ article, the question is far from resolved.
However, contrary to what this letter to the editor implies, it's a
problem of _too many_ viable hypotheses (i.e. a symptom of true
science), rather than an inability to come up with viable hypotheses.

DEATH:
  [See _The Evolutionary Biology of Aging_, by Michael Rose]
  Accidental death:  for evolution to bother to _try_ to explain
accidental death would be silly.  It is simply a part of the natural
world.
  Aging/Senescence/"Natural" death:  Evolution can, indeed, explain
this.  As a result of the fact that accidental deaths occur, the
probability of living to age X decreases with increasing X (even with
a constant probability of accidental deaths).  If the constant
probability of accidental death within an age class is 0.1, for
instance, then the probability of living from age class 0 to age class
1 is 1 - 0.1 = 0.9.  The probability of living from age class 0 to age
class 2 is (1 - 0.1)^2 = 0.81, the probability of living to age class
3 is (1 - 0.1)^3 = 0.729, etc.  Because the probability of surviving
to a given age decreases with increasing age, selection against
deleterious effects later in life is weaker than selection earlier in
life.  Thus, either (1) traits which cause beneficial effects early in
life but deleterious effects later in life will be selected _for_, or
(2) deleterious mutations which don't act until later in life will not
be strongly selected _against_, and will come into equilibrium at a
higher frequency in the population.  I'd offer again to show the math,
but I got no takers the last three times, and I'm gonna be gone for a
week anyways (e-mail me if you want to see it, and I can post it when
I get back).

--
--Andy  (adpeters@sunflower.bio.indiana.edu)