Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Species without successful genome inheritance

1 view
Skip to first unread message

Michael Young

unread,
Jun 12, 2010, 11:19:57 AM6/12/10
to
I remembered when I applied to an internship in California about some
organism that reproduced sexually and, when it did, it produced
progeny that did not have the successful genetic properties of its
parent, but had a seemingly random genome. I was able to actually find
the short description from the professor studying it:

"Many colonial/clonal marine organisms exhibit an extraordinarily
diverse array of life-history strategies, some of which appear to
contradict evolutionary theory. For example, an entire class of
bryozoans (class Stenolaemata) reproduce sexually via polyembryony:
the production of a single zygote which divides up to several hundred
times to produce a clone of offspring. The cloned genotype is a
unique, unproven genetic combination, rather than a copy of the mother
successful genotype. This strategy is similar to xeroxing a single
lottery ticket hundreds of times in order to win, and appears to make
little sense. Understanding this strategy may throw light on the
evolution of sex. [...] We have 2 years of data that suggests there is
no climax state in these communities, and the pattern of succession
appears to be highly seasonal. There are a large number of introduced
species (approximately 35% of the fauna on our experimental panels),
and many are dominant space competitors."

Any thoughts on this? It's interesting because it seems to defy
evolutionary theory. How does randomly mixing up the genome of your
progeny every generation contribute to more successful proliferation?
Just thinking about it myself, I suppose if you do that, you may
always have a number of your progeny that will survive, and this
reproductive method might make them extremely adaptable to an
environment that changes quickly or to a disturbance event.

But it seems inefficient...to me it's like an analogy of serving in
tennis. You always throw the ball and swing the racket a certain way
to get the ball in the service box. When you establish a good throw
and swing and stance, you learn to stick with it, so you have a higher
chance of getting the ball in. This reproductive method is like
hitting the ball with a random tossing style, stance, and swinging
form every time, and if the ball goes in the box, great, but there is
a refusal to memorize how the ball got there, and the randomness
continues. It doesn't seem to make much sense...

Michael Young

unread,
Jun 12, 2010, 11:27:31 AM6/12/10
to
silly me, forgot to link reference...

http://www.humboldt.edu/~hsureu/mentors.html#craig

marks...@yahoo.com

unread,
Jun 12, 2010, 12:03:10 PM6/12/10
to
<snip>

The cloned genotype is a unique, unproven genetic combination, rather
than a copy of the mother successful genotype. This strategy is
similar to xeroxing a single
lottery ticket hundreds of times in order to win, and appears to make
little sense.
<snip>
The above bit makes no sense to me . If each copy has a unique genetic
makeup then it is not at all like copying a lottery ticket. It would
be like getting a pile of tickets all with different sets of numbers,
and that would help a win.

Since I dont understand half the big words in that reference I cant
comment further.

Michael Young

unread,
Jun 12, 2010, 12:11:09 PM6/12/10
to
On Jun 12, 12:03 pm, marks542...@yahoo.com wrote:

> The above bit makes no sense to me . If each copy has a unique genetic
> makeup then it is not at all like copying a lottery ticket. It would
> be like getting a pile of tickets all with different sets of numbers,
> and that would help a win.

I completely agree, he could have used a better analogy (which is why
I tried to make my own; dunno how successful it really is).

r norman

unread,
Jun 12, 2010, 1:14:03 PM6/12/10
to

I don't understand why there is any problem at all. The real question
is why Prof. Craig
http://www.humboldt.edu/~hsureu/mentors.html#craig
is making such a big deal out of ordinary biology.

OK, I do understand why -- that is how you take an interest in an
obscure group and try to make it sexy and important enough to get
grant money and students and all.

Still, here is what is going on. The Bryozoa are a group of animals
that ordinarly live colonial lives where all the members of a colony
are clonally derived from a specific individual. The members of the
colony, the zooids, can be somewhat distinct functionally, and are
often not completely independent individuals. See, for example,
http://en.wikipedia.org/wiki/Bryozoa

So what happens is that an individual reproduces sexually to produce
an offspring. This new individual has a "unique, unproven genetic
combination, rather than a copy of the mother successful genotype" as
Prof. Craig comments, but that is true of EVERY sexually reproducing
organism.

This one individual then divides asexually (to form a clone) of zooids
that forms the colony. The unusual thing about this particular class
of Bryozoa is that the fertilized egg (the zygote) divides mitotically
(asexually) to form the colony. Most colonial multicellular organisms
are formed by asexual reproduction from a multicellular parent, not
from a single celled parent.

So the evolutionary 'problem' is that the genome gets replicated to
form many individuals before ever being subject to selection (except
insofar as the cell is capable of functioning). I don't see that as
any particular problem. In fact you can say that a multicellular
organism does exactly that -- it reproduces the "untested" new genetic
package by cloning to form a tremendous number of cells that can then
go out into the world to try its skill at reproduction and survival.
These thingies are pretty much doing exactly the same thing except
that instead of forming a single multicellular organism from a zygote,
it forms a colony of multicellular organisms from the zygote.


John Harshman

unread,
Jun 12, 2010, 1:51:49 PM6/12/10
to
marks...@yahoo.com wrote:
> <snip>
> The cloned genotype is a unique, unproven genetic combination, rather
> than a copy of the mother successful genotype. This strategy is
> similar to xeroxing a single
> lottery ticket hundreds of times in order to win, and appears to make
> little sense.
> <snip>
> The above bit makes no sense to me . If each copy has a unique genetic
> makeup then it is not at all like copying a lottery ticket. It would
> be like getting a pile of tickets all with different sets of numbers,
> and that would help a win.

No, what happens here is that the bryozoan has sex, once, producing a
single zygote with a single genotype (half from each parent), and then
the zygote divides many times to produce hundreds of identical zygotes.
So it is indeed like copying a single lottery ticket.

But I have to wonder: how many reproductive zooids does a single colony
have? Just one? Unlikely. So I bet the colony is producing a bunch of
offspring of different genotypes.

John Harshman

unread,
Jun 12, 2010, 1:49:03 PM6/12/10
to
Michael Young wrote:
> I remembered when I applied to an internship in California about some
> organism that reproduced sexually and, when it did, it produced
> progeny that did not have the successful genetic properties of its
> parent, but had a seemingly random genome.

Sort of. In the same way you don't have the successful genetic
properties of your mother, but have a seemingly random genome, i.e. half
from your father.

> I was able to actually find
> the short description from the professor studying it:
>
> "Many colonial/clonal marine organisms exhibit an extraordinarily
> diverse array of life-history strategies, some of which appear to
> contradict evolutionary theory. For example, an entire class of
> bryozoans (class Stenolaemata) reproduce sexually via polyembryony:
> the production of a single zygote which divides up to several hundred
> times to produce a clone of offspring. The cloned genotype is a
> unique, unproven genetic combination, rather than a copy of the mother
> successful genotype. This strategy is similar to xeroxing a single
> lottery ticket hundreds of times in order to win, and appears to make
> little sense. Understanding this strategy may throw light on the
> evolution of sex. [...] We have 2 years of data that suggests there is
> no climax state in these communities, and the pattern of succession
> appears to be highly seasonal. There are a large number of introduced
> species (approximately 35% of the fauna on our experimental panels),
> and many are dominant space competitors."
>
> Any thoughts on this? It's interesting because it seems to defy
> evolutionary theory. How does randomly mixing up the genome of your
> progeny every generation contribute to more successful proliferation?

That's the problem of sex. The only weird thing here is the massive
polyembryony. There are other groups that do it too, like various gall
wasps. The reasons for sex are endlessly debated, and you might want to
look up the literature. Your question really has little to do with
polyembryony; just sex.

Ron O

unread,
Jun 12, 2010, 2:31:29 PM6/12/10
to

Well, to start off with there is no such thing as an extant species
that does not have a successful genome inheritance.

You also have the incorrect assertion that the progeny inherit genomes
that are less well adapted or in some way deficient compared to the
parent. If this were true the species would have been extinct long
ago. The progeny are, obviously, on average, sucessful enough to
insure the species survival.

The particular embryo production process isn't that strange or
deficient than others that are common in the animal kingdom. Just
think of all the organisms that produce only one egg or progeny (such
as humans) at a time. This is like producing one offspring, but this
species throws in r/K selection. Instead of putting a lot of effort
into producing one child this species, just makes more copies of that
one offspring and sends them off to fend for themselves. It is a
strategy that obviously works for that species.

Why it doesn't make a bunch of different embryos to disperse is likely
due to the species reproductive history. The ancestral species might
have only produced one embryo at a time, and the clonal divisions is
just a stop gap means to make more embryos without messing with the
existing meiotic process. There might also be a benefit to producing
clones for dispersal. They might combine and form more viable
colonies where ever they end up attaching. You'd have to look at
their population structure and biology to figure something like that
out. As for why it is better to combine the genetics from two
individuals instead of sticking with what works, what do you think
evolution means? One of the worst things that a species can do is
become genetically uniform and be adapted to only one specific niche.
They do fine until things change. In the case of this species where
their parents are living and the species competition that the parents
are dealing with is not only constantly changing, but the progeny are
not likely to settle in exactly the same environment with the same
species as neighbors. Being a little different may help and obviously
does, and the multiple copies of the same animal give that genetic
type more opportunities to settle in a place that is suited to that
genotype.

In the end organisms are stuck with what they inherit from their
ancestors. They can change things to some extent with new mutations
or combinations of genes, but somethings are very difficult to change
without screwing everything up. The meiotic process and embryo
formation for this species is likely something like that. Look at
humans. We are stuck with a pretty stupid meiotic process. It worked
great when our ancestors were the size of mice and rats and reproduced
every 8 weeks, but we stall out the egg cells at meiosis I after the
first meiotic division. The egg cells have to spend their time in
purgatory until they get the signal to continue the meiotic division
to produce haploid cells. In the case of humans the egg cells have to
spend years waiting to complete cell division. By the time a human
female is 40 years old over 1% of her eggs lose or gain chromosomes
because of messed up chromosomal segregation, and it just gets worse
with age until menopause. It would be nice if we could snap our
fingers and tell the cells to complete meiosis and then wait to
mature, but we are stuck with a very important process that evolved
under different conditions.

http://en.wikipedia.org/wiki/R/K_selection_theory
http://en.wikipedia.org/wiki/Meiosis

Ron Okimoto

r norman

unread,
Jun 12, 2010, 3:05:43 PM6/12/10
to

It doesn't matter how many reproductive zooids a colony has if they
are all clones. In that case, it doesn't matter whether they are
produced from a single single-celled zygote or from a multi-cellular
zoid except in for the fact that the zoid has undergone some selection
if it survives long enough to reproduce asexually.

Before you complain about that last point, it seems completely
irrelevant to me but it has to do with the "untested" nature of the
genome that gets copied asexually.

This situation is really no different from any organism that shows
both sexual and asexual reproduction. The reason for using both
techniques and the evolutionary strategy for determining just when in
the life cycle and to what extent each reproductive method gets used
is, indeed, an interesting question in biology. That, no doubt, is
what the faculty member really means.

Paul Ciszek

unread,
Jun 12, 2010, 5:40:01 PM6/12/10
to

In article <2d6bf386-a330-4abd...@z8g2000yqz.googlegroups.com>,

Michael Young <young...@gmail.com> wrote:
>
>Any thoughts on this? It's interesting because it seems to defy
>evolutionary theory. How does randomly mixing up the genome of your
>progeny every generation contribute to more successful proliferation?

It is not random random, it is a random selection taken from the two
sets of genes that each of the parents received from their parents.
All sexually reproducing creatures do this. Since the parents lived
long enough to have sex, they have been selected for success to some
degree. ("Lived long enough to have sex" may sound like a pretty
lame critereon by human standards, but for many creatures that is
quite a challenge.) Overall, this has proven to be a very effective
way to maintain diversity while promoting genes that are beneficial
for survival; almost every multicellular organism on Earth reproduces
sexually.

In short, how these insects produce a zygote
is no different then how almost all other sexually reproducing
creatures produce a zygote; the only difference is that they then
produce a whole bunch of offspring from that same zygote, rather
than producing a bunch of slightly different zygotes (more commonly
called "siblings"). This is unusual, and it certainly sounds like
a stupid idea to me, but then, swimming upstream and dying after
breeding only once also sounds like a pretty stupid idea to me,
and yet it is done.

Because evolution is not a directed, "designed" process, it produces
a lot of stupid features and dead ends. Over time, most of these
get weeded out. Some set of past circumstances that no longer apply
can lead to an entire species being burdened with the same stupid
feature; if it is not bad enough to outright prevent survival, it
may persist for a long time simply because the genes for *not* having
the stupid feature do not exist anywhere in the population. It can
even be re-inforced in a perverse way--for example, an individual
salmon that decided *not* to swim all the way upstream and die, but
rather pick a nice place a little ways upstream, would be left without
anyone of the same species to spawn with. Similarly, one of these
insects that had a gene for not doing the clone-thing may not be able
to reproduce with "normal" mates, or might have fewer offspring.` It
can be hard to change the way a species reproduces, so silly stuff
gets perpetuated. Peacock's tail and all that.

--
Please reply to: | "The anti-regulation business ethos is based on
pciszek at panix dot com | the charmingly naive notion that people will not
Autoreply is disabled | do unspeakable things for money." -Dana Carpender

Michael Young

unread,
Jun 12, 2010, 6:28:53 PM6/12/10
to
Thanks for the responses, which is exactly why I posted it here to
begin with: Dr. Craig's presentation of the species made it sound to
me like something it wasn't. R. Norman did a wonderful job clearing up
the air, and I now understand what's actually going on here. Ron O
brings up the fact that the species would have gone extinct long ago
if the genome was, as Paul puts it, "random random," which is what I
thought Craig was telling us. Turns out the description was a tad
misleading, and now I know that really what's going on is normal
sexual reproduction with a proliferation of clonal colonies as an
aside. In that regards, essentially, the species is like any other
normal sexually-reproducing species undergoing natural selection.

I don't know whether it's my fault that I missed this, by
miscomprehending what Dr. Craig said, or whether he deliberately made
it sound like something really unusual. Either way I'm kind of
irritated, but thanks to you gentlemen for pointing me in the right
direction.

John wrote:

> Your question really has little to do with
> polyembryony; just sex.

Well I wasn't even thinking of sex at that point. I understand how
evolution works, it being my favorite biology subject, but I thought
what was going on here wasn't your normal mix-up of genes upon sexual
reproduction; I thought this species was reproducing and for some
reason making progeny that had purely random genomes regardless of the
success of the parents' genomes, which doesn't make any sense, and I
couldn't figure out how that could work in nature. Now I know that's
not actually happening here.

Thanks for filling me in.

Ron O wrote:

> You also have the incorrect assertion that the progeny inherit genomes
> that are less well adapted or in some way deficient compared to the
> parent.

Right, that's why I was confused, I think I misunderstood the
description.

John Harshman

unread,
Jun 12, 2010, 9:25:58 PM6/12/10
to
r norman wrote:
> On Sat, 12 Jun 2010 10:51:49 -0700, John Harshman
> <jhar...@pacbell.net> wrote:
>
>> marks...@yahoo.com wrote:
>>> <snip>
>>> The cloned genotype is a unique, unproven genetic combination, rather
>>> than a copy of the mother successful genotype. This strategy is
>>> similar to xeroxing a single
>>> lottery ticket hundreds of times in order to win, and appears to make
>>> little sense.
>>> <snip>
>>> The above bit makes no sense to me . If each copy has a unique genetic
>>> makeup then it is not at all like copying a lottery ticket. It would
>>> be like getting a pile of tickets all with different sets of numbers,
>>> and that would help a win.
>> No, what happens here is that the bryozoan has sex, once, producing a
>> single zygote with a single genotype (half from each parent), and then
>> the zygote divides many times to produce hundreds of identical zygotes.
>> So it is indeed like copying a single lottery ticket.
>>
>> But I have to wonder: how many reproductive zooids does a single colony
>> have? Just one? Unlikely. So I bet the colony is producing a bunch of
>> offspring of different genotypes.
>
> It doesn't matter how many reproductive zooids a colony has if they
> are all clones.

Yes it does, if they're doing sexual reproduction. Which is the claim here.


John Harshman

unread,
Jun 12, 2010, 9:27:02 PM6/12/10
to
Paul Ciszek wrote:
> In article <2d6bf386-a330-4abd...@z8g2000yqz.googlegroups.com>,
> Michael Young <young...@gmail.com> wrote:
>> Any thoughts on this? It's interesting because it seems to defy
>> evolutionary theory. How does randomly mixing up the genome of your
>> progeny every generation contribute to more successful proliferation?
>
> It is not random random, it is a random selection taken from the two
> sets of genes that each of the parents received from their parents.
> All sexually reproducing creatures do this. Since the parents lived
> long enough to have sex, they have been selected for success to some
> degree. ("Lived long enough to have sex" may sound like a pretty
> lame critereon by human standards, but for many creatures that is
> quite a challenge.) Overall, this has proven to be a very effective
> way to maintain diversity while promoting genes that are beneficial
> for survival; almost every multicellular organism on Earth reproduces
> sexually.
>
> In short, how these insects produce a zygote

Insects? We're talking about bryozoans here.

r norman

unread,
Jun 12, 2010, 9:36:20 PM6/12/10
to
On Sat, 12 Jun 2010 18:25:58 -0700, John Harshman
<jhar...@pacbell.net> wrote:

Telll me what is the difference between two zooids that are clones
each reproducing sexually and a single zoid reproducing sexually two
times in quick succession?

What is the difference between producing sperm (or eggs) from your
testis (or ovary) on the left side and producing sperm (or eggs) from
your testis (or ovary) on the right? And what is the difference if
you have a whole series of different testes or ovaries? (You actually
do but they are all crammed into a pair of packages.) And what is the
difference if you put those testes lr ovaries into separate zooids?
Remember, they all have the same diploid genome.

John Harshman

unread,
Jun 13, 2010, 12:18:06 AM6/13/10
to

None. But there would be considerable difference between one zooid
reproducing twice and two zooids reproducing twice. Why are the multiple
zooids given lesser capabilities than the single zooid in your scenario?

> What is the difference between producing sperm (or eggs) from your
> testis (or ovary) on the left side and producing sperm (or eggs) from
> your testis (or ovary) on the right? And what is the difference if
> you have a whole series of different testes or ovaries? (You actually
> do but they are all crammed into a pair of packages.) And what is the
> difference if you put those testes lr ovaries into separate zooids?
> Remember, they all have the same diploid genome.

Not getting the point here. Now, I was supposing that each zooid
reproduced once. If they reproduce many times, so much the better.

r norman

unread,
Jun 13, 2010, 8:48:12 AM6/13/10
to
On Sat, 12 Jun 2010 21:18:06 -0700, John Harshman
<jhar...@pacbell.net> wrote:

You are right -- I did not get the point at all. You said " So I bet


the colony is producing a bunch of offspring of different genotypes"

and I focused on the "different genotypes" instead of the "bunch".

The main advantage of producing a clonal colony like this is that you
have many times the reproductive structures and so can potentially
produce many times the number of offspring. However there is no
difference in the different genotypes produced this way.


Walter Bushell

unread,
Jun 13, 2010, 2:02:55 PM6/13/10
to
In article <hv0urh$sdn$1...@reader1.panix.com>,
nos...@nospam.com (Paul Ciszek) wrote:

> ("Lived long enough to have sex" may sound like a pretty
> lame critereon by human standards, but for many creatures that is
> quite a challenge.)

Actually for most of the history of man, surviving long enough to have
sex has been a challenge. Surviving infancy was major, even for hunter
gatherers, who were otherwise healthier than agriculturalists.

--
All BP's money, and all the President's men,
Cannot put the Gulf of Mexico together again.

Robert Carnegie: Fnord: cc talk-origins@moderators.isc.org

unread,
Jun 13, 2010, 3:44:53 PM6/13/10
to
Ron O wrote:
> Well, to start off with there is no such thing as an extant species
> that does not have a successful genome inheritance.

That seems a rather dogmatic statement, but you make a... long...
argument that if there is such a thing, it ain't this one.

I can't address this otherwise except by noting that a significant
proportion of individual pieces of published science are, in the end,
wrong, and so it would take more than one study of "The Beast that
Disproved Darwin" to overturn serious scientific opinion.

r norman

unread,
Jun 13, 2010, 3:53:29 PM6/13/10
to

This is by no means an example of evolution producing a stupid feature
and leading to a dead end. There are innumerable creatures that mix
sexual with asexual reproduction. The sexual stage, of course,
produces genetic variation while the asexual stage, of course,
produces lots of copies of the same genotype.

If you are already pretty well adapted to your envrionment and the
environment is quite stable, then an excellent strategy is to make
more copies of yourself rather than trying all sorts of new stuff. We
think of evolution as "change" and sometimes conclude, therefore, that
change is necessarily good. However stability is more often the
wisest course.

I am speaking, of course, of biological evolution. Politically I
rather favor well thought out and wisely chosen change. OK, in most
cases, ANY change is better than keeping what we've got!


Michael Young

unread,
Jun 13, 2010, 5:27:36 PM6/13/10
to
On Jun 13, 3:53 pm, r norman <r_s_nor...@comcast.net> wrote:

> OK, in most
> cases, ANY change is better than keeping what we've got!

Or so we believe, until the change winds up worse off than what we had
before. In some cases, at least.

Robert Carnegie: Fnord: cc talk-origins@moderators.isc.org

unread,
Jun 16, 2010, 4:47:51 AM6/16/10
to

The trouble with political change is, it's done by politicians -
whether a mob setting fire to the palace or tearing down the prison is
also involved, or not.

I think in 2010 what still matters most is who has control of the
television stations. So you may want to consider who's running them
just now.

0 new messages