In the hopes of benefitting Jim, lets define some other terms which
might help him to understand how biologists think about transitions.
The first term is "intermediate".  An intermediate organism is one
that contains features that are used to classify two different groups
of organisms (I'll discuss that more in a moment), or that contains a
feature that is intermediate (along some scale) between the homologous
features in two other groups of organisms.  For example, feathers are
a defining feature of birds, while a single row of undifferentiated
teeth on each jaw is a defining feature of reptiles (turtles may
disagree :-).  Since _Archaeopteryx_ has both of these traits, it is
an intermediate form.  If an intermediate is located in the proper
strata, then it can be postulated that the intermediate is also a
transitional organism.  This means that the features shared with one
group are shared because the transitional organism descended from
members of that group, while the features shared with the other group
are shared because the transitional organism is an ancestor of the
animals of that second group.  In the above case, _Archaeopteryx_ is
believed to contain undifferentiated teeth because it descended from
other organisms (i.e.  reptiles) that had that feature.  Modern birds
have feathers because they are descended from an organism (perhaps
_Archaeopteryx_) that acquired that characteristic.

The upshot of all of this is that we can *hypothesize* that
_Archaeopteryx_ is a transitional organism between reptiles and birds.
The hypothesis is subject to refutation.  If it passes many possible
tests, we'll gain enough confidence to call _Archaeopteryx_' status as
a transition between reptiles and birds a fact.  Currently the
evidence in this case is equivocal.  Why is that?  Because the
hypothesis of transitional status is a good scientific hypothesis,
meaning that there are many tests it *can* (read must) pass in order
to gain acceptance.

So what are some of those tests?  Let's start with one that's
relatively easy to understand and that I've hinted at above.  Does the
organism's existance occur at the right time?  One of the reasons that
_Archaeopteryx_'s status is equivocal is that there are some fossils
older than _A_ which appear to contain features which are not
possessed by _A_, and yet which are possessed by modern birds.  If it
becomes well established that there are modern bird features possessed
by earlier animals, then we can rule out _A_'s designation as
transition between reptiles and birds.  Going in the opposite
direction, if there are reptilian features that _A_ does not possess,
and that are retained in modern birds, the hypothesis will have been
similarly refuted.

As promised, I'm now going to ramble about features and
classification.  Chris is always telling us about the hierarchical
pattern in which you can classify organisms.  Let's dissect this a
bit.  Unfortunately I think that we have to begin with some
vocabulary.  A most important concept in evolutionary biology is
polarity analysis.  Polarity analysis is the comparison of features
for the purpose of distinguishing the ancestral condition from the
derived condition.  Continuing with my previous example, for birds the
ancestral condition is a beak full of teeth.  The derived condition is
a beak without teeth.  The ancestral character is referred to as
plesiomorphic or a plesiomorphy.  The derived condition is referred to
as apomorphic or an an apomorphy.  Please note that polarity analysis
implies *no* connotations of progress.

Another set of definitions that you need here deal with phylogeny.  A
monophyletic grouping is a grouping of organisms that contains an
ancestral form and all of its descendents.  A paraphyletic grouping is
one which contains an ancestor, and some of its descendents.  A
polyphyletic grouping is one which contains groups of organisms that
are derived from different ancestors.  These definitions are described
with the help of a figure below:


       ^
       |     A   B   C     *  D  
     t |       \/   /       \/    
     i |       a \/   *    /      
     m |          b\ /   /    *   
     e |            x \/    /     
       |               c\ /       
       |                 d\       
       |                    \     
       |                      e  
       |                        
       |__________________________

The *'s represent extinct lineages added for realism :-) Capitol
letters are organisms alive now, small letters are organisms that were
alive at some point in the past.  A grouping which consists of A,a,B
and b is paraphyletic, because it leaves out C, which is also one of
b's descendents.  A grouping which includes A,a,B, and D is
polyphyletic because it lumps A,a, and B which are derived from b with
D which is derived from c.  A grouping which includes the whole chart
is monophyletic because it consists of e and all of e's descendents.

So how do you make a tree like the above?  The answer is to use
polarity analysis to determine characters which can be used to define
groups.  Apomorphies which define groups are called synapomorphies.
Characteristics which exist in one group, but also in other groups
outside of the group in question are called symplesiomorphies.  To get
an idea of how the nomenclature is used, feathers are a synapomorphy
defining the clade aves, but are a symplesiomorphy shared between, for
example, sparrows and starlings.  This gets back to the hierarchy that
Chris is always trying to hammer home; as you move back towards the
root of the evolutionary tree, characteristics that are
symplesiomorphies within the group become synapomorphies defining the
(more inclusive) group.  Once you get to the part of the tree where
the characteristic is a synapomorphy, no organism outside of that
group should have that trait (e.g. no animal outside of aves has
feathers).

When Chris says that characteristics fit into a hierarchical pattern,
he means that comparative studies allow us to choose characteristics
which meet the definition of synapomorphy above, and which can thus be
used to generate a tree consistent with the idea that all traits arive
from modifications of traits possessed by previous members of the same
lineage.  I suspect that with this background, Jim can now accept that
a transitional organism can be defined as an organism in which a
synapomorphy first appears (and all of its descendents will thus form
a monophyletic group).  Of necessity, said organism also contains
characteristics used to define previously existing groups (e.g.
undifferentiated teeth), and hence it will also be an intermediate.
An intermediate need not be transitional, however.  For example, in
the diagram above, C might be intermediate between B and D (or more
importantly, between b and D), but its intermediacy is predicated on
the basis that it too is derived from c, rather because it is a part
of the lineage connecting B and D.

Mickey Rowe     (rowe@pender.ee.upenn.edu)