Roughgarden's evolution of sex paper in Am Nat (Reply)

Doug.Eernisse at UB.CC.UMICH.EDU Doug.Eernisse at UB.CC.UMICH.EDU
Sun Apr 12 17:47:28 EST 1992

Chris Colby asked for some response on a model of the evolution
of sex by Roughgarden in Am. Nat. volume 138.
I do not claim to have studied Roughgarden's model in any
detail, and only have now glanced at it prompted by a number of
inconsistencies in Colby's posting. Here are my $0.02.
>         The standard reason given for the evolution of sex is that
> it allows groups to evolve faster. Beneficial mutations occuring
> in separate organisms could be shuffled onto the same chromosome
> via recombination (whereas sequential mutations would have to occur in
> an asexual population). Thus, the benefit of sex (to a group) was tied
> to the ability to recombine; this is especially useful when
> the environment is unstable. Sex is assumed not to be adaptive to
> individuals.
Without defending this conventional argument, one does not have
to cast it in such a group selectionist manner. The individuals
who happened to possess the more successful combinations would
certainly gain in their inclusive fitness (as would their
>         Roughgraden attempts to decouple sex from recombination and
> develops a one-locus model of why sex may be adaptive to groups of
> organisms. He shows that in a sexual population, the distribution of
> genotypes stays more stable with time (because random mating brings
> about genotype frequencies in Hardy-Weinberg eq.) than in an asexual
> population when the fitnesses of the different genotypes fluctuate.
> So, when the fitnesses of the genotypes change, there are always more
> representatives of the most favored genotype in a sexual population
> than in an asexual one. The benefit of sex increases (in his model)
> with the magnitude of environmental fluctuation.
>         (His model is one-locus, two allele. The asexual population
> is diploid and hermaphroditic. The sexual population is (of course)
> diploid and randomly mated.
This seemed a bit odd to me. This implied that the sexual
population was gonochoric (dioecous), later becoming
hermaphroditic while reproducing asexually. Why should an
asexual organism produce any sperm? In
glancing at the article in question, I found the following
statements which seemed even more troublesome:
p. 938: "...we first investigate a completely promiscuous
mating system, then a completely monogamous system, and finally
a system in which promiscuity and monogamy are mixed. In all
cases we focus on a simultaneous hermaphrodite, that is, an
organism with both male and female gonads. This is probably the
most common type of individual that exists, seeing that most
individual flowering plants produce both seeds and pollen. Also,
a great many invertebrates, such as barnacles and snails, have
both male and female gonads. Dioecy, where each individual has
only a single sex, can presumably be considered a derivation of
simultaneous hermaphroditism whereby a parent, rather than
packaging both male and female gonads within its body, instead
produces males that function as physically detached extensions
of itself at the cost of deferring male function until the male
can mature."
What!?! Simultaneous hermaphrodites are not the most common
type of individual, although they are certainly common among
land plants. Among metazoans, which he appears to be mainly
referring to by his emphasis on sperm production, simultaneous
hermaphroditism is characteristic of only a few "major"
lineages, e.g., barnacles, the so-called "higher" gastropods,
some bivalve families, ctenophores, a flatworm-gnathostomulid
?clade, and tunicates (I probably missed some examples). Based
on outgroup comparison, simultaneous hermaphroditism is likely
derived from gonochorism in each of these cases. Roughgarden has
worked on barnacles at Hopkins Marine Station, near his home
base at Stanford. This is one of the premier examples of a
diverse rocky shoreline assemblage of organisms. He should know
better. Hopkins >used to be< a center for that sort of
knowledge, including a former student Michael Ghiselin who wrote
the 1969 Q. Rev. Biol. article "The evolution of hermaphroditism" 
and the 1974 book "The Economy of Nature and the Evolution of Sex"
(neither cited) in which he carefully considered the
phylogenetic distribution of hermaphroditism. His perception of
deriving gonochorism (dioecy) in any general sense from a
hermaphroditic condition is clearly out of touch with reality.
What Roughgarden means by "monogamous" is also a bit peculiar.
p. 940. "[Monogamous mating is defined as when] an individual is
fertilized by, and fertilizes, only one other individual.
Couples form at random but, once formed, remain intact."
Roughgarden then uses modeling to conclude that under such
circumstances, natural selection should maximize egg production
with each hermaphrodite only producing enough sperm to fertilize
its partners eggs. This is contrasted with "promiscuous" mating
where sperm-sperm competition leads to a equal investment in
eggs and sperm and a twofold cost of meiosis. I guess his point
is to try to argue that most mating systems are "mixed" since
they lie somewhere between these extremes, so a twofold cost of
meiosis is never realized.
Is he really this naive? First, why did he never even mention
the possibility of self-fertilization? This is not asexual
reproduction of course, but his monogamous mating model is
simply (IMHO) an odd way to reformulate similar conclusions
reached by the Charlesworths and others who have shown that
regular selfers should produce tiny amounts of sperm or pollen.
I cannot think of anything in nature that would come close to
his "monogamous mating" system. However, there are plenty of
selfing plants and in freshwater and terrestrial habitats,
selfing molluscs are not uncommon. 
p. 951: "Monogamy ameliorates the cost of meiosis. The finite
dispersal capabilities of organisms entail that they mate
monogamously more than is possible in a completely mixed
population. Thus, the cost of meiosis should rarely, if ever, be
a full halving of the species' growth rate. Indeed, data
summarized by Lloyd (1988) show that allocation to female
function in flowering plants usually exceeds 50%, which implies
a less-than-maximal cost of meiosis."
Again, plants frequently self so it is not surprizing that the
female allocation, on average, exceeds 50%. Lloyd's compilation
does not appear to support his argument.
In the sea, regular selfers are rather rare. Only about 20 or so 
metazoan species have been shown to regularly self-fertilize. 
Almost all of these are almost certainly derived from gonochoric 
ancestors, since they are the exceptions in otherwise gonochoric 
clades (see Strathmann et al. in Am. Nat. 123: 796-818). Their
gonads contain mostly ovaries and only tiny bits of testes. These 
are not to be confused with "major lineage" cases of simultaneous 
hermaphroditism discussed above, whose members typically possess
often elaborate mechanisms to ensure cross fertilization and 
have nearly equal allocation to eggs and sperm.
Perhaps not incidently (since I was provoked to write so much),
I have published on both chiton and bivalve simultaneous
hermaphrodites which are among the handful of marine metazoan
species that appear to normally reproduce without cross 
fertilization (Biol. Bull. 174: 287-382 and Biol. Bull. 175:
218-229, respectively, in the latter case with coauthor Diarmaid 
O'Foighil). I mention these only because the latter case
has proven quite interesting. At the time we wrote our article,
we thought that clam in question, in the genus Lasaea, could be
either selfing or parthenogenetic, but we favored the selfing
hypothesis because my coauthor had elsewhere published his
observations of a sperm penetrating a presumed normal egg,
followed by extrusion of normal appearing polar bodies.
Subsequent cytogenetic studies by O'Foighil and coworkers
have now shown that Lasaea probably instead undergoes
apomictic (asexual) parthenogenesis, no meiosis occurs and the
sperm never fuses with the egg's nucleus, and this all appears
to be somehow connected with polyploidy (these Lasaea have
perhaps the highest ploidy levels known for molluscs). To my
knowledge, this is the first known case of self-sperm activated
parthenogenesis. If anyone knows of other cases, I would be
grateful. Sperm-activated parthenogenesis is not uncommon in
certain lizards and fish, for example, but these examples are of
hybrid origin and I think always depend on sperm from one of the
parent species, not sperm they produce themselves. As in Lasaea,
polyploidy is typically involved. Anyway, Diarmaid should have
some mtDNA sequence data to report before too long on these
interesting clams (bringing this back to molecular evolution),
and Roughgarden's model from a simultaneous hermaphrodite to an
asexual diploid condition seems a bit esoteric.
Perhaps Roughgarden's model does indeed point to a previously
ignored potential benefit of sex (i.e., resistance to change
rather than ability to change). My criticism only involves
his presumptions about reality. Meanwhile, I will continue
to investigate situations in which organisms appear to do
quite well without sex or at least without cross-fertilization
and, based on phylogenetic comparison, I can establish that their 
condition is derived. With enough of these examples perhaps we can
address the issue of when sex no longer is adventagous and so 
get at the factors that more normally maintain it as the rule.
| Doug Eernisse         Doug_Ee at                  |
|                       userlegd at umichub.bitnet                  |
| Museum of Zoology, Univ. of Michigan, Ann Arbor, MI 48l09 USA  |
| Phone: (313) 747-2193  Fax: (313) 763-4080                     |

More information about the Mol-evol mailing list