Responses from David Wake

Question 1.

Name: Group 4, Kellar's Fri. 11:00 Dis
        Rachel, Lisa, Sheila, and Scott
Subject: Constraints 

Do taxa with richer and lenghtier histories necessarily evolve less than 
younger taxa that are less "constrained?"  Does a group become more 
historically constrained as it progresses and evolves or doesn't evolve 
through time for the simple fact that anything that did or didn't happen 
before it directly impact what it is presently.  On a bigger scale, shall 
we expect to see less evolution in all taxa as time progresses as they 
all get temporally further from the earliest unicellular common 
ancestors?

Response to Question 1.  I fear that we might be dealing with taxonomic
artifact.  It is true that some ancient lineages (e.g., lungfish) persist
but do not speciate, whereas other lineages that are far more recent (e.g.,
fishes of the family Cichlidae or Percidae) have speciated greatly.  There
are some basal actinopterygian fishes (the larger clade to which teleost
fishes such as cichlids and percids belong) that also are very poor in
numbers of species (for example, the families to which the bowfin, Amia,
belongs), but what you see is that we just split off the ancient groups to
one side and keep progressing upwards toward the large "progressive"
groups.  If you carry this far enough, in a cladistic diagram you would
have  a basal Amia as a sister group of all teleosts, and a basal lungfish
as a sister group of all terrestrial vertebrates.  We are just lucky to
have a few of these basal groups still hanging on!  Perhaps another way of
looking at it is to ask if we have ever had any basal lineages that have
just hung on for millions and millions of years and then started
speciating.  I cannot think of any examples, but there may be some.




Question 2.

Larry Rabin, Tina Bratis, Roger Liu
2-3pm Steve D.
Group #5

Professor Wake,
Our question is on the evolution of plethodontid salamanders- How does 
the evolution of miniturization and a large genome size compete with one 
another. If an animal has a large genome, does that not mean it also have 
large cells and thus a larger body? How does the evolution of 
miniturization effect this?
Also, is the projectable tongue of chameleons similar in development 
compared to the highly projectable tongue of Hydromantes?  Are these 
tongues considered convergent? Thank you for a very thought provoking class.

Response to Question 2.  
        In general one expects that species with large genomes will have
large cells and that they will also have large bodies.  This is exactly the
case with modern lungfishes, especially the South American and African
species, which have the largest vertebrate genomes.  However, interesting
things happen when large genomes are pushing up from below, so to speak,
and community dynamics are pushing down from above.  This seems to be
happening in some species of plethodontid salamanders in California and
Mexico.  For example, the genus Thorius includes the smallest tetrapods
(they become sexually mature at about 16 mm head+body length -- about 60%
of an inch!  These animals have vary large genomes (ca 30 pcgm DNA/haploid
genome, versus less than 10 in any mammal and usually around 5) and so they
face a real crunch.  They have completely gotten rid of the cerebellum in
the brain and their skull is very weak and incompletely formed.  It is sort
of shrunken down around the edges of the critically important sense organs
- nasal capsule, eyes and inner ears.  There is a huge fontanelle (an
opening uncovered by bone) over the brain, and the forebrain
(telencephalon) is shrunken.  There are only about 5% glial cells in the
brain (versus about 50% in mammals) and there are many peculiar features of
the skeleton (these were described by a former graduate student from my
group - James Hanken, currently a professor at the University of Colorado -
in the Journal of Morphology and the Biological Journal of the Linnean
Society in 1983-84).  Hanken and I reviewed some of the factors involved in
biological miniaturization in an article in the 1993 Annual Reviews of
Systematics and Ecology.  We pointed out that when organisms have very
large cells it might be useful to rethink what we mean by biological size
-- the very small Thorius are actually far smaller than they appear to be
in a morphogenetic sense, because they have to make complicated vertebrate
structures such as limbs and brains, with very few cells.
        There are only superficial similarities between the tongues of
plethodontid salamanders and chameleons.  In chameleons the tongue slides
off a single tapered rod (a bone, a part of the hyobranchial appratus) that
lies in the center of the floor of the mouth.  This rod is elevated and
aimed as the mouth is opened, and there is a very powerful ring muscle that
wraps around it.  This ring muscle contracts and the entire tongue is
propelled rapidly toward the prey.  Whether it hits the prey or misses, the
tongue then just fall, by gravity, and swings back and forth in the air. 
It is then slowly and laboriously cranked back into the mouth by several
different muscles.  The chameleon tongue system has been recently studied
in very great detail in three beautiful papers published in the J. Exp.
Biol. 1992, by our earlier consultant, Al Bennett, and a postdoc, Peter
Wainwright.  The most extreme salamander tongue is long, but far less long
than the chameleon tongue.  However, it is much faster and more accurate. 
It is associated with two tapered rods in the floor of the mouth known as
the hyobranchial apparatus.  Whereas in the chameleon the hyobranchial
apparatus remains in the floor of the mouth and the tongue is shot off of
it, in plethodontids a pair of circular muscles wrap around tapered
cartilaginous rods at the hind end of the  apparatus.  Now the entire
hyobranchial apparatus, with the sticky tongue connected to it anterior
end, is shot from the mouth, and the circular muscles remain in the head! 
It is exactly the reverse of the chameleon.  Furthermore, the retractor
muscles are fired at the same time as the protractor (the circular)
muscles, so that the tongue flies to the prey and is being pulled back as
it goes, so that when it reaches it maximum extension it immediately
reverses and flies back to the mouth.  It is much faster than the
chameleon, both on the way out but especially on the way back.  Some
plethodontids can hit the prey in about 6-7 msec after muscular activity is
first detected in the circular muscle.  My work on salamander tongues
started with papers in the Journal of Morphology (with former student R. E.
Lombard) in 1976 and 1977, and my most recent paper (on kinematics) was
done with Larsen and Beneski in J. exp. Zool. (1989).




Question 3.

Names:          Michael Feinberg and Mehrdad Afrahi
Group Number:   Four (4)
T.A.:           Steve Deban
Section:        Friday, 2-3 p.m.

Topic:          Evolutionary Constraints

Our question is whether constraints result in directionality of evolution?
For example, Archaeopteryx had developed feathers primarily for insulation;
then, feathers were exapted for flight.  In present time, the most abundant
bird species are birds of flight.  Is it possible that the development of 
feathers resulted in a constraint, thus keeping birds of flight, as well as 
flightless birds, from better adapting to their environment through means 
other than flight?  Had there not been any feathers, would flight have 
been an option?

 Response to Question 3. 
        Several authors have observed that what start out in evolution as
novelties sometimes turn into constraints.  It is a little hard for me to
think of feathers as being constraining, though.  There are lots of birds
that have lost flight -- for example, rails on oceanic islands.  There are
also birds that have become aquatic  or semiaquatic (penguins, auks) and
use feathers only as insulation.  It is unlikely that birds will ever
become burrowing and wormlike, but that is probably related to things in
addition to morphological features related to flight.  Sometimes
constraints at one level become opportunities at another.  I have written
about one of these, the loss of lungs in salamanders.  Most salamanders
that lose lungs (and it has happened in parallel in at least five separate
clades) live in and around rapidly flowing streams (lungs seem to be of
negative value in such conditions; they act like water wings and bouy the
animals up, causing them to lose their grip on the substrate and be swept
downstream.  These animals have excellent cutaneous circulation and
exchange gases through the skin, so the lungs are not missed as far as can
be determined).  Most of these lineages include extreme stream-adapted
specialists that have not done much.  But in one group the loss of the
lungs seems to have freed up the lung-filling mechanism and it has been
coopted to become a superb  tongue-projection mechanism.  This group
subsequently underwent a relatively enormous amount of speciation, though
cause and effect is very difficult to sort out.




Question 4.

my question concerns methodology.  i am interested to know how the use of
a mitotic inhibitor such as colchicine would reveal the presense of homo-
plasy.  wouldn't inhibiting mitosis result in tetraploid nuclei and thus
coding more information than found in the normal diploid state rather than
showing the reversion to a bauplan?
stacey leann smith

Response to Stacey Smith's question (#4) about colchicine.  The surprising
thing about colchicine as used with amphibians is that mitoses appear more
or less normal, except for the fact that the process is greatly slowed
down.  The cells all remain diploid.  Thus, one can mimic the effects of
large genomes and features that are characteristic of late ontogeny do not
appear.  The technique works better if it is applied topically to local
areas, such as limb buds or even regenerating  limbs.




Question 5.

Topic: Sympatric Speciation
Group: 2 at 2-3 with 
        Steve
Kevin and Gary

We were particularly interested in your work with the ring group
of salamanders.  Are there any other in depth studies of possible
sympatric speciation that you would recomend reading?

Response to Question 5.

        I do not consider the Ensatina case to be one of sympatric
speciation, but rather of a particular kind of allopatric speciation --
isolation by distance with restricted gene flow leading eventually to
differentiation of terminal groups.  Sympatric speciation has been studied
in recent years mainly by two groups -- one associated with Guy Bush at
Michigan State University (working mainly with true fruit flies that switch
host plants - for example from native hawthorns to introduced and
domesticated apples, and from native sour cherries to introduced and
domesticated sour and sweet cherries), and the other with the two Drs.
Tauber of Cornell Univ. (also entomologists -- they recently edited a book
on sympatric speciation).  The main judgement of people about sympatric
speciation is that it can happen, but is rare and not very important, and
this certainly is Futuyma's opinion.  I tend to be a bit more lenient.  It
could be important in some groups, but I do not think it happens at all in
groups like birds, mammals and herps, although cases of what Bush (see
recent review in TREE) calls alloparapatry may SEEM like sympatry to some
folks.




Question 6.

Topic: Macroevolutiom
>From Ib 160 group 3 (F 2-3)
Jacques Finlay
Christina Campbell
Christoph von Pohl
David Cyranoski
Rasmus Nielsen

Dear Dr. Wake,
In many peoples view, micro evolution in a neo-Darwinian sense,
and speciation can explain the observed patterns of morphological 
diversification.  In your lectures as well as in the article, 
you argue for the use of concepts such as Rstep functionsS in 
evolutionary biology.  In the lectures, you used the apparent discrete 
division of the RmorphospaceS into separate units as an argument for 
applying such explanations.  However, the existence of discrete 
morphological units is a direct prediction of neo-Darwinian theory.  
The absence of gene flow between units, and the structural 
relationship between units, can cause such a pattern.  
Furthermore, you argue for the incorporation of RlimitsS into 
evolutionary theory.  Can such limits not be described simply by 
correlations in fitness and absence of genetic variation?  We would 
like to know why it is necessary to evoke mysterious concepts such 
as Rstep functionsS when the existing theory is able to explain the 
observed pattern of morphological evolution?

Resonse to question 6. 
        We may be dealing with an issue in "ways of seeing" here.  First I
am totally unconvinced that speciation has anything to do with
morphological diversification, or anything, expect putting stop-points in
phylogenesis.  I know that many microevolutionists like to assign
speciation a role in morphological diversification, but I will have none of
it!  Speciation may or may not involve morphological differentiation.  To
me it is simply a red-herring.  
        So, now onto the main point of the question.  Step -functions.  I
do not think of them as mysterious at all.  I also have trouble with the
idea that gaps in morphospace are related to absence of genetic variation,
and also disagree with the idea that fitness relations and patterns of
genetic variation and genetic variance-covariance relations predict gaps in
morphospace.  They will not predict specific gaps, which in fact can be
predicted from knowledge of developmental patterns and phylogenies.  Have I
ever used the term "step-function" in print?  I think not.  Perhaps I did
in lecture.  I do like to talk about thresholds, which as I understand it
is orthodoxy both in quantitative genetics and in developmental genetics. 
This is a very large question to respond to quickly, but basically I see
many examples of evolution where there are sharp limitations on variation
that are seemingly inexplicable based on gaps in genetic variation, but
become clear in a developmental and phylogenetic context.  The evolutionary
transition from 5 to 4 toes in frogs and salamanders, as contrasted with
lizards, is one.  In lizards there is a gradual shortening of  a toe until
it is lost, and this can be understood within the context of standard
theory as Lande has shown well (1976, Evolution).  However, I know of no
selection theory that predicts the abrupt change from 5 to 4 toes within a
species, and even assymmetrically within an organism.  Furthermore, I know
of no selection theory that predicts that in one taxon the lost toe is
number 5 in lineage after lineage, whereas in the other taxon it is number
1, again in lineage after lineage.  However, combined knowledge of
development (yes, developmental constraints) and phylogeny will do it.  I
have a paper dealing with these issues in press in Evolution (with Shubin
and Crawford as co-authors). It should appear in early 1995.




Question 7.

Name: Dave Kilimnik, David Richards, Misty, Shanti, Han.
Topic: Exaptation

Professor Wake: Do you believe that exaptation is fundamentally 
structuralist or functionalist?  Take feathers for example.  Feathers may 
have evolved from structures that originally functioned for 
thermoregulation.   There must have been natural selection when they were 
coopted for flight.  This would appear to be functionalism.  But they 
became feathers, not extra limbs or ejectable spears.  This would appear 
to be structuralism, i.e. evidence of design limitations.  Or is it just 
absence of natural selection for extra limbs?  Can you shed some light?  
Thanks.

Response to question 7.  
        I think exaptation is fundamentally functionalist in perspective. 
The idea is that natural selection "latches onto" whatever is handy. 
Structuralism has more to do with fundamental rules of morphogenesis and of
transformation from one form to the next (for example, from a triangle to a
square by adding a new structural element, rather than introducing angles,
breaking an element and adjusting lengths).  Structuralism has more to do
with rules of form organization and transition (e.g., fundamental changes
in ontogenetic trajectories), than using existing materials.  Tinkering is
basically a functionalist concept -- but one I like!  It is pretty closely
related to exaptation.




Question 8.

TA: Steve Deban
Section Time:Fri, 12-1
Group #3: Adam Norman
          Teresa Chung
          Colleen Carrero
Question: Is it possible that developmental and structural options 
increase?  How much do constraints put on design novelty? and Do complex 
cells increase in complexity?

Response to question 8.
        I am not sure I understand the question.  First, I do like to think
about constraints, but more as a crutch than anything else.  I see a bias
in the production of variation.  Some workers find it convenient to think
that just the right kind of variation and selction have not yet coincided. 
I think in general that historical (i.e., phylogenetic and developmental)
and formal (i.e., physical and architectural) constraints do limit what can
happen and bias what does along certain lines.  Constraints are not
absolute in their effects.  For example, insects seem constrained to
produce bodies enclosed in chitin and composed of three main kinds of
segments, but they do not seem to suffer for it!  They are most unlikely to
produce bone, and even less likely to produce bony internal skeleton for
appendages.   As to the cell question -- I do not know.  I am uncomfortable
with the classification system that we use for cells, and we might be
seeing a kind of "taxonomic" artifact.  Do complex cells give rise to even
more complex cells?  It is a kind of developmental question, since cells,
like organisms, are not "born " in one cell division, but arise through a
series of ontogenetic tranformations.  I suppose that complex cells might
undergo one more cell division to become even more complex, or simpler, but
I just do not have the empirical background to say.




Question 9.

Names:  Dane Post, Duncan Parks, Pamela Wong
Subject:  Molecular Phylogeny

In your paper in the reader, you say that morphological data exhibit a 
lot of homoplasy.  Would you support the exclusive use of molecular data 
for phylogeny reconstruction, given this homoplasy?  Do you think 
molecular data are more reliable for reconstruction?

Response to question 9.

        In the first place, homoplasy is an outcome of a phylogenetic
analysis.  Morphological and molecular data alike display homoplasy, and if
anything molecular data show vastly more homoplasy than morphological data.
 There are two reasons.  First, taxonomists quickly abandon morphological
data that is too variable and so they high-grade in favor of "good"
characters (i.e., those that are known, a priori, to show differences among
taxa).  With molecular data, and especially sequence data, homoplasy is so
common that very rarely does a worker even attempt to apply so-called
cladistic principles -- recognition of  plesiomorphic and apomorphic
states.  Instead one used unordered characters.  In general it seems to me
that we get closer to the one "true" phylogeny with more characters, even
though with more characters we inevitably get more homoplasy.  I advocate
working with morphology and trying to understand it so that we "high-grade"
morphology not just from a biased view of what is more or less variable but
from an understanding of morphogenesis and morphological transitions (I
could give you lots of examples; some are in my article).  Then I would
analyze all data sets separately and do consensus  analysis to see where
they all agree.  Then I would combine data in a "total evidence" approach. 
We are very far from having all data, and so one might think that this is a
blind man and elephant approach, but the fact is that we get a more or less
standard vertebrate phylogeny from many separate data sets, from fragments
of data sets and also from combinations of all data.  So, maybe we are
getting close after all, even in spite of all the problems.
        I almost forgot.  One nice thing about mapping morphologies on a
molecular data based tree and then comparing it with a morphologically base
tree is that you can examine the homoplasies in morphology carefully and
based on morphogenetic principles decide whether to choose the most
parsimonious tree, or one that is almost as parsimonious and costs very
little more.   Of course, all of our phylogenies should be viewed as
hypotheses to be tested by further data.




Question 10.

Names:  Kevin Davey, Derek Hitchcock, Scott Turner, Jean Yim
        Kellar 2-3, Group 4
Topic:  Synthesizing functionalism and structuralism

When thinking about the concepts of structuralism and functionalism, it is
immediately evident that both play a significant role in evolution and the
diversity of life.  Therefore a synthesis of the two makes sense.  We are
unclear, however, as to how to begin this synthesis, as we see the two
concepts so divergent initially.  Could you give some ideas as to how a
functionalist and structuralist can come together and participate in a
useful discourse?

 Answer to Question 10.
        Not an easy question!  I have tried to exemplify an approach in my
1991 Amer. Nat. paper.  Fundamentally, this is a dialectical approach.  You
do the best you can with each approach, recognize fundamental conflicts,
and then try to achieve a synthesis on the basis of dealing with the
conflicts one by one.  See my answer to question 6, where I provide a start
of an answer.  I argued that numbers of toes is more readily understood as
an outcome of miniaturization and knowledge of developmental patterns than
as an adaptive response to selection, and this was a structuralist
interpretation, but at the same time I recognized that miniaturization
almost certainly was best understood within the framework of selection and
the working out of community dynamics.  I was challenged on this by Reeve
and Sherman (1993, Quar. Rev. Biology), who in a fiercely and exclusively
functionalist argument insisted that direct selection was a priori the
explanation of choice for an evolutionary biologist and should only be
abandoned once all attempts to show it have been exhausted.  Thus, they
would have me study these miniature animals in detail and prove in case
after case after case that there was no selective advantage to having four
toes before I invoked what is in essence a side-effect, structuralist
hypothesis.  Obviously I disgree with them, but they do represent a point
of view that is still widely prevalent in the field.




        That it it!  Thanks for the challenging questions.  I hope that you
enjoyed the VDG. 

        I will see you at the final exam, and wish you all well.   David