AMER. ZOOL., 40:789–800 (2000)
Developmental Data and Phylogenetic Systematics:
Evolution of the Vertebrate Limb1
PAULA M. MABEE2
Department of Biology, University of South Dakota, Vermillion, South Dakota 57069
INTRODUCTION
Contributions of phylogenetic systematics
Virtually every hypothesis regarding the
evolution of development can be made
more exact by a fuller incorporation of phylogenetic methods and comparative morphology. Phylogenetics provides tools for
explicitly and rigorously developing devo/
evo hypotheses. In this paper I point out the
contributions that phylogenetic systematics
has made thus far to evolutionary developmental biology, and I describe the specific methods that are necessary for devo/
evo research. My primary objective is to
show by example why phylogeny, morphology, and development are necessary for
generation of rigorous devo/evo hypotheses.
1 From the Symposium on Evolutionary Developmental Biology: Paradigms, Problems, and Prospects
presented at the Annual Meeting of the Society for
Integrative and Comparative Biology, 4–8 January
2000, at Atlanta Georgia.
2 E-mail: [email protected]
Phylogenetic reconstructions or hypotheses, at all levels, for virtually all organisms, are the primary contributions from
phylogenetic systematics to the synthesis
between development and evolution. A revolutionary change in methods of phylogenetic reconstruction began in the 1960s
with the publication of Hennig’s work Phylogenetic Systematics (1966). Hennig laid
out the principles of phylogenetic reconstruction and their connection to evolution
with unusual clarity (Hull, 1988). He demonstrated that once characters have been
properly sorted out with respect to phylogeny, they form hierarchically ordered or
‘‘nested’’ groups. Descent with modification (and retention of modifications) leads
to this hierarchical ordering of shared derived similarities among organisms. Thus,
in contrast to much previous systematic
thought, overall similarity was shown not
to be a guide to evolutionary relationship.
Only shared derived features provide evi-
789
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
SYNOPSIS.
Among the primary contributions of phylogenetic systematics to the
synthesis of developmental biology and evolution are phylogenetic hypotheses. Phylogenetic hypotheses are critical in interpreting the patterns of evolution of developmental genes and processes, as are morphological data. Using a robust phylogeny, the evolutionary history of individual morphological or developmental features can be traced and ancestral conditions inferred. Multiple characters (e.g.,
morphological and developmental) can be mapped together on the phylogeny, and
patterns of character association can be quantified and tested for correlation.
Using the vertebrate forelimb as an example, I show that by mapping accurate
morphological data (homologous skeletal elements of the vertebrate forelimb) onto
a phylogeny, an alternative interpretation of Hox gene expression emerges. Teleost
fishes and tetrapods may share no homologous skeletal elements in their forelimbs,
and thus similarities and differences in Hox patterns during limb development
must be reinterpreted. Specifically, the presence of the phase III Hox pattern in
tetrapods may not be correlated with digits but rather may simply be the normal
expression pattern of a metapterygium in fishes. This example illustrates the rigorous hypotheses that can be developed using morphological data and phylogenetic
methods.
‘‘Creating a general reference system and investigating the relations that extend
from it to all other possible and necessary systems in biology is the task of systematics.’’ (Hennig, 1966, p.7)
790
PAULA M. MABEE
Phylogenetic hypotheses are starting
points for comparative evolutionary
studies
Many developmental and evolutionary
biologists have pointed out the critical role
for phylogenetics in the synthesis between
developmental and evolutionary biology
(Wake et al., 1991; Coates, 1994; Hadfield
et al., 1995; Raff, 1996; Wray, 1996; Shubin et al., 1997; Tabin et al., 1999). Holland
and co-workers, for example, have consistently used phylogenies in interpreting the
patterns of evolution of genes and developmental processes (e.g., Holland and Garcia-Fernandez, 1996; Wada et al., 1998).
They point out that ‘‘Unless phylogenetic
relationships between living organisms are
known, comparative developmental biology
can give limited insight into the evolution
of developmental mechanisms’’ (Holland
and Garcia-Fernandez, 1996, p. 389). That
phylogenetic hypotheses underpin interpretations of experimental research was demonstrated by Coates (1994) in his excellent
review of limb development and evolution.
He pointed out, for example, that the developmental genetic explanation of the
prevalence of pentadactyly assumes tetrapod monophyly. In another example, Ahlberg and Milner (1994) noted that the largescale cladistic review of sarcopterygians
was a conceptual turning point in the study
of tetrapod origins. In their review of the
evolution of Hox genes and animal origins,
Knoll and Carroll (1999) used the phylogeny of metazoans to determine the ancestral
and derived conditions of a number of characters. Clearly, the importance of phylogenies is understood in studies of evolution
and development.
Developmental biologists, for the most
part, have led the intellectual synthesis between evolution and development. Arthur
(1997, Fig. 12-1), summarizes the various
disciplinary contributions towards an overall understanding of evolution in a pie diagram, yet there is no slice representing
systematics even though it plays a central
role. Arthur, however, clearly understands
the importance of phylogenies in interpreting the evolution of developmental mechanisms (Arthur et al., 1999). His oversight
is perhaps a sign of the success of phylogenetics: trees are such a basic part of the
landscape that they are taken for granted.
Trees have been rapidly incorporated into
several disciplines over the past decade, and
new comparative subdisciplines, such as
evolutionary ecology, have been spawned
(Brooks and McLennan, 1991). The incorporation of trees into the devo/evo synthesis
is important and necessary as illustrated
further in this paper.
A phylogenetic tree provides an explicit
framework from which a specific hypothesis regarding the evolution of any characteristic may be inferred. For example, if a
scientist were interested in the evolution of
a character such as ‘‘number of Hox
genes,’’ the number of Hox genes would be
mapped onto the tree and optimized at ancestral nodes in order to infer a single hypothesis (or set of hypotheses) for the pattern of Hox evolution (e.g., Holland and
Garcia-Fernandez, 1996; Knoll and Carroll,
1999). Phylogenetic methods for reconstructing ancestral character states, also
known as character optimization or character mapping techniques, have been central in ‘‘getting many nonsystematists to
think about phylogeny and evolutionary
history’’ (Omland, 1999, p. 604). Essentially, these methods allow an investigator
to infer the pattern of evolutionary history
of a particular feature. Reconstruction of
ancestral states is crucial to a wide range of
research programs in evolutionary biology.
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
dence of common ancestry, and thus primitive and derived similarities must be distinguished in order to reconstruct trees. For
example, if one were interested in reconstructing the evolutionary relationships
among amphibians, only features that
evolved within amphibians would be relevant; other characters would be uninformative. Because the notochord, a dorsal rod
of tissue that underlies the neural tube, is
found in all developing chordates (a more
inclusive group than amphibians), it is not
useful in reconstructing the evolutionary
history within amphibians. It is a shared
primitive feature of amphibians. It is, however, a shared derived feature of chordates
and as such is useful in hypothesizing their
recent common ancestry.
DEVELOPMENT
AND
Theoretical example
A theoretical example that mirrors an actual example of character optimization
(‘‘Vertebrate limb evolution,’’ below) is as
follows: Consider a monophyletic group
consisting of eight taxa for which the phy-
791
FIG. 1. Hypothetical example of character optimization.
logenetic relationships have been reconstructed as in Figure 1. For simplicity, one
might imagine that this phylogeny has been
constructed using characters entirely independent of the ones to be mapped (this is
unnecessary: see Maddison and Maddison
[1992]). Presence or absence of each of
three characters is mapped at the branch
tips; taxon names are not. The two most
basal taxa, as well as basal members of
clade 1, possess characters A, B, and C.
Two terminal sister taxa within clade 1 have
only characters A and B. All members of
clade 2 have only character C. The most
parsimonious optimization of these characters is as indicated: Character C is lost in
the most recent common ancestor of the
two lineages in clade 1 that have only A
and B. Characters A and B are lost in the
most recent common ancestor of clade 2
(Fig. 1). Therefore, the two terminal sister
taxa in clade 1 share no derived homologies
with the taxa in clade 2.
Vertebrate limb evolution
The study of the vertebrate limb is one
of the classic anatomic and evolutionary
examples since Owen’s day, and it is an increasingly celebrated ‘‘model system’’ of
comparative developmental genetics (Nelson et al., 1996; Shubin et al., 1997; Arthur
et al., 1999). In 1930, the vertebrate anatomist Goodrich stated: ‘‘Few questions
concerning the general morphology of vertebrates have aroused greater interest [than
the limbs]’’ and this remains true today.
Forelimb morphology of representative
gnathostomes is diagrammed on the branch
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
The most commonly used technique for
reconstruction of ancestral states is based
on the criterion of parsimony (Maddison
and Maddison, 1992). This involves choosing the historical reconstruction that involves the fewest evolutionary changes
(Farris, 1970, 1983; Fitch, 1971) for discrete (vs. continuous) characters. However,
there may be a number of equally parsimonious reconstructions for a particular
character (Swofford and Maddison, 1992),
and all need to be considered when the goal
is to infer the underlying processes. Maddison and Maddison (1992) provide an excellent summary of methods for reconstructing patterns of character evolution.
Some examples of studies in which authors
have reconstructed the pattern of character
change using a phylogeny, and then made
inferences about evolutionary processes include: Huey and Bennett (1986); Maddison
(1990); Brooks and McLennan (1991); Harvey and Pagel (1991); Sanderson (1991);
and Donoghue et al. (1998). The apparent
simplicity of mapping a character onto a
phylogeny, however, is belied by the complexity of assumptions and challenges of
character optimization, which have been
discussed by Swofford and Maddison
(1992); Frumhoff and Reeve (1994); Maddison (1994); Ryan (1996); Schultz et al.
(1996); Cunningham et al. (1998); Cunningham (1999); and Omland (1999)
among others. The questions that interdisciplinary biologists pose about the evolution of developmental features are similar
in many respects to those posed by comparative evolutionary biologists from other
disciplines. These include: 1) What is the
pattern of character evolution? 2) Are particular characters correlated evolutionarily?
3) What are the probabilities of different
types of character change, such as loss and
gain? Patterns of character evolution can be
analyzed to answer these and other questions (Maddison and Maddison, 1992).
PHYLOGENETICS
792
PAULA M. MABEE
tips of a phylogeny for the jawed vertebrates, i.e., the gnathostomes (Fig. 2). Gnathostomes are considered the sister taxon of
lampreys (Forey and Janvier, 1993; Janvier,
1999). The phylogeny of gnathostomes is
fairly well resolved and is based primarily
on morphological data. Among recent vertebrates, the gnathostomes include the basal
Chondrichthyes (sharks, rays, chimeras),
which is sister to the actinopterygian fishes
plus sarcopterygians (lungfish, coelacanths,
and tetrapods). The land vertebrates (tetrapods) are a monophyletic group within the
sarcopterygians.
The evolution and details of morphology
of the limb endoskeleton of gnathostomes
are well synthesized by Coates (1994) and
Coates and Cohn (1998). Here I add additional data on actinopterygian limb morphology, with special attention to basal actinopterygian pectoral fins. I optimize homologous forelimb skeletal elements on a
phylogeny for gnathostomes (Fig. 2), and
hypothesize ancestral conditions and derived states.
A ‘‘tribasal’’ pectoral fin, made up of a
propterygium, mesopterygium, and metapterygium (Fig. 3) is considered the primitive condition for living sharks and rays
(neoselachians) in some (Compagno, 1973;
Shirai, 1996) but not all (Maisey, 1984)
systematic studies of chondrichthyans. Determination of the ancestral condition for
chondrichthyans is dependent upon the
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
FIG. 2. Phylogeny of gnathostomes. Relationships within Actinopterygii follow Lauder and Liem (1983) and
within Sarcopterygii follow Maisey (1986). Two major extinct gnathostome clades, basal to recent taxa listed
above include the placoderms, sister to gnathostomes and acanthodians (not shown), the sister-group of sarcopterygians 1 actinopterygians (Janvier, 1981). Gardiner (1984) concludes that Polypterus is the sister to Chondrosteans (represented by the paddlefish, Polyodon, on Fig. 2) 1 neopterygians (gars, Amia, and teleosts).
Diagrams of forelimbs at branch tips are (from left to right): Chondrichthyes (represented by Squalus [adapted
from Jarvik, 1980]); Teleosts (represented by Salmo [adapted from Goodrich, 1930]); Polyodon (representing
Acipenseriformes); Polypterus (adapted from Goodrich, 1930]); coelacanth (adapted from Ahlberg, 1992); lungfish (represented by Neoceratodus [adapted from Goodrich, 1930]); Euthenopteron (adapted from Jarvik, 1980)
and tetrapods (represented by hindlimb of Ichthyostega [adapted from Ahlberg and Milner, 1994]). Tetrapods
express Hox gene phases I, II, and III, and teleosts express phases I and II (see text).
DEVELOPMENT
AND
placement of some basal fossil taxa. Because this is unresolved, it is uncertain
whether the tribasal condition is in fact the
primitive condition for chondrichthyans (de
Carvalho, 1996). The tribasal pectoral fin is
considered primitive for chondrichthyans
herein; I describe an alternative optimization below. The conclusions drawn regarding homology and Hox genes are not dependent on a primitive tribasal condition for
gnathostomes.
Basal actinopterygians possess a pectoral
fin with propterygial and metapterygial elements; the homology of the mesopterygium is less well understood, but middle
mesopterygial radials (Coates, 1994) are
present (Fig. 4). Given this tribasal condition in basal actinopterygians and chondrichthyans, it is most parsimonious to consider it primitive at the level of gnathostomes
(Fig. 2). Basal actinopterygians thus may
be interpreted to have retained the primitive
tribasal condition from the most recent
common ancestor shared with sarcopterygians and chondrichthyans (Fig. 2). Consistent with this argument, Coates (1994) suggests that the propterygium of actinopterygians is a retained primitive feature.
Teleost fishes have lost the metapterygium (Coates, 1994; Coates and Cohn,
1998). Though Goodrich (1930, p. 156)
noted that ‘‘the (posterior) axis in the Teleostei has disappeared and all the radii
come to articulate on the girdle,’’ he did not
specify that it was the metapterygium that
was lost. Similarly, though Rosen et al.
(1981, p. 204) interpret the actinopterygian
793
FIG. 4. Posterolateral view of the left pectoral fin of
Polyodon spathula, 48.3 mm standard length. Radials
1 and 2 correspond to the mesopterygium, and radials
3–6 correspond to the metapterygium. Radial 6 is still
developing, and may be postaxial.
pectoral fin as a ‘‘transformation from a
primitive (chondrichthyan) metapterygial
fin into a propterygial type,’’ they do not
explicitly specify the loss of the metapterygium as a shared derived feature of Teleostei. The propterygium, present in all teleosts (Fig. 5), is interpreted here as a retained primitive feature (Fig. 2). It is sometimes misidentified as the distal radial one
(Sordino et al., 1995; Grandel and SchulteMerker, 1998). Correct identification of the
morphological homologies of the pectoral
fin is critical, however, in correctly interpreting the evolution of Hox gene expression patterns. The mesopterygium is a little
mentioned part of the pectoral fin endoskeleton. It refers to the middle of the three
proximal parts of the tribasal fin (Fig. 3),
and in actinopterygians it may be consid-
FIG. 5. Lateral view of the pectoral fin of the teleost,
Danio rerio, the zebrafish.
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
FIG. 3. Tribasal limb morphology that may be the primitive condition for gnathostomes. Shown is Squalus,
adapted from Jarvik (1980).
PHYLOGENETICS
794
PAULA M. MABEE
well studied. In this optimization, as in the
previous, teleosts are distinguished by the
loss of the metapterygium. Sarcopterygians,
however, are not characterized by the loss
of the pro- and mesopterygium, but primitively retain the absence of these elements
and retain the presence of a metapterygium.
Teleosts and tetrapods share no skeletal
homologues in the pectoral fin
According to either optimization, teleosts
and tetrapods share no skeletal homologues
in the pectoral fin. That is, the teleost fin
has a propterygium and mesopterygium,
and lacks a metapterygium, but the tetrapod
limb is comprised of a metapterygium, and
not a pro- or mesopterygium. There are no
shared skeletal components with this interpretation.
Hox expression in tetrapod limb
development
Many genes are critical to limb development, but the vertebrate Hox genes are of
great interest because of their importance in
limb patterning. In the tetrapod limb (chick
and mouse models), the Abd-B related
genes of the HoxD cluster are expressed in
a complex, dynamic pattern encompassing
at least three distinct (Nelson et al., 1996),
independently regulated (Gerard et al.,
1993; van der Hoeven et al., 1996) phases
(Shubin et al., 1997). In the earliest phase,
phase I, HoxD-9 and HoxD-10 are expressed across the entire limb bud (Nelson
et al., 1996). This expression correlates
with the time that the stylopod (the most
proximal element of the forelimb, e.g., the
humerus) is specified (Nelson et al., 1996).
Phase II, correlated with the development
of the zeugopod (e.g., the forelimb radius/
ulna) is initiated in response to Sonic
hedgehog. Hox genes are expressed in a
nested set centered around the Sonic expressing cells, with HoxD-13 being expressed in the most restricted domain, and
HoxD-12 and HoxD-11 each encompassing
a broader domain (Nelson et al., 1996). The
latest phase, phase III, is correlated with the
autopod (digits). In phase III, HoxD-11–13
are expressed across most of the distal portion of the limb bud (Nelson et al., 1996).
The expression pattern in phase II ‘‘revers-
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
ered homologous to those radials between
the propterygium and metapterygium. In
acipenseriforms (Fig. 4), and other basal actinopterygians, the number of mesopterygial radials varies. The four proximal radials that are primitive for teleosts (Johnson
and Patterson, 1996) are considered homologous to the mesopterygium.
Sarcopterygians have lost the propterygium and mesopterygium; they retain exclusively metapterygial skeletons (Rosen et
al., 1981; Coates, 1994; Carroll, 1997) (Fig.
2). Within sarcopterygians, tetrapods are
further characterized by the presence of
digits (Shubin and Alberch, 1986; Ahlberg
and Milner, 1994). Digited, polydactylous
limbs originated after tetrapods diverged
from their shared ancestry with lungfish,
but before the evolutionary radiation of living forms (Coates and Clack, 1990; Coates,
1994). Digits are generally interpreted as
postaxial branches off the preaxially arched
metapterygial axis (Shubin and Alberch,
1986; Ahlberg and Milner, 1994) (Fig. 2).
Given these assumptions regarding primitive conditions and character homologies,
the above optimization may be summarized
as follows: the ancestral gnathostome possessed a tribasal pectoral fin, which was retained in some chondrichthyans and all basal
actinopterygians (Fig. 2). Within actinopterygians, teleosts lost the metapterygium; in
the common ancestor of sarcopterygians, the
propterygium and mesopterygium were lost
(Fig. 2).
Because the ancestral gnathostome condition cannot be hypothesized with certainty, an additional character optimization
must be considered. If chondrichthyans
primitively possessed a metapterygial pectoral fin (Maisey, 1984), the tribasal condition (i.e., the pro- and mesopterygium)
must have evolved independently within
chondrichthyans and again in the most recent common ancestor of actinopterygians.
That the propterygium would then be considered an independently acquired feature
of actinopterygians is consistent with the
study of Jessen (1972), whose work was
followed in subsequent systematic studies
(Rosen et al., 1981; Patterson, 1982; Gardiner, 1984; Maisey, 1984, 1986). The evolution of the mesopterygium has not been
DEVELOPMENT
AND
es’’ such that HoxD-13 has the broadest expression domain and HoxD-12 and HoxD11 are nested within it (Nelson et al., 1996).
These expression patterns are reviewed in
Shubin et al. (1997).
Alternative interpretation
An alternative interpretation of the difference in phase III Hox expression, however, emerges from the mapping of morphological features and Hox gene expression together on the gnathostome phylogeny (Fig. 2). Specifically, the ‘‘unique’’
phase III pattern of tetrapods may have
795
nothing to do with the evolution of digits
(found only in tetrapods), but may instead
relate to the presence of a metapterygium,
primitively retained in tetrapods from the
common ancestor of gnathostomes. The
presence of the phase III Hox pattern in tetrapods thus may simply be the normal expression pattern of a metapterygium.
Additionally, the interpretation that phase
I and II Hox expression patterns are correlated with the compartments of the stylopod
and zeugopod, respectively (Shubin et al.,
1997) must be revised. Teleosts lack a metapterygium and yet demonstrate Hox phases I and II. Given the likely complete lack
of homology between forelimb skeletal elements of tetrapods and teleosts, the proposed correlation between their shared early
gene expression patterns (phases I and II)
and the identity of skeletal elements must
be reevaluated.
A test of two interpretations
As a test of whether Hox phase III expression is uniquely correlated with digits,
the metapterygium, or neither, the expression pattern should be examined in a gnathostome with the primitive tribasal condition. Either a basal gnathostome such as a
shark or a basal actinopterygian such as a
sturgeon would be equally instructive. The
fins of both lack digits yet possess pro-,
meso-, and metapterygia. If phase III is present in either of these taxa, it is clearly not
uniquely associated with digits, but rather
with the presence of a metapterygium. A
basal sarcopterygian such as a lungfish, presenting a metapterygium and no digits also
could be examined for phase III Hox expression. Similarly, if phase III is present,
it is clearly not uniquely associated with
digits, but rather with the presence of a metapterygium.
Nelson and Tabin (1995, p. 631) recommended examining the Hox expression of
fin development of a primitive fish as well
as closer relatives (lungfish) to ‘‘test the
novelty of the autopod.’’ Although they indicate that this would be a stronger test of
the hypothesis that the unique Hox pattern
represents digits, they do not present a specific alternative hypothesis or interpretation.
Explicit tests of hypotheses regarding char-
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
Hox expression in teleost (Danio) limb
development
In the teleost, Danio rerio (zebrafish),
Hox expression in phases I and II is similar
to that observed for tetrapods (Sordino et
al., 1995). This is interpreted as a reflection
of the conservation of early developmental
mechanisms (Sordino et al., 1995). However, the latest phase of Hox expression in
tetrapods (phase III) is not observed in the
zebrafish; the expression of the HoxD-11,
12, and 13 cognate genes is not extended
anteriorly (preaxially) into the limb bud
(Sordino et al., 1995).
This difference in late HoxD expression
pattern between the tetrapod and teleost is
interpreted as a ‘‘molecular illustration of
the absence in teleosts of a structure homologous to the tetrapod distal autopod
(digits)’’ (Sordino and Duboule, 1996, p.
118). In other words, phase III is correlated
with the presence of digits (Sordino and
Duboule, 1996). Similarly, Shubin et al.
(1997) suggest that the presence of phase
III Hox expression in tetrapods but its absence in teleosts indicates that the pattern
may be apomorphic for tetrapods or a more
inclusive group. Moreover, they state that
the separate single cis-regulatory enhancer
for phase III Hox expression in tetrapods is
‘‘more derived relative to conserved phase
I and II enhancers.’’ Shubin et al. (1997)
further propose that the shift from preaxial
(zeugopod) to postaxial (autopod 5 digits)
branching is correlated with the ‘‘reversal’’
in phase III HoxD gene nesting relative to
that in phase II, and that this is correlated
with the fossil record.
PHYLOGENETICS
796
PAULA M. MABEE
acter association such as this will be formulated given a phylogenetic framework
and additional morphological data.
Uses of developmental data in systematics
Phylogenetic systematists are interested
in how data from development can be used
to reconstruct phylogenies. Information on
polarity, order, and homology are critical in
phylogenetic reconstruction, and all have
been hypothesized to be available from ontogeny (Nelson, 1973, 1978; Roth, 1984,
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
DISCUSSION
Developmental biologists have traditionally sought explanations for conserved
structures or ‘‘universal’’ mechanisms during embryonic development (Bolker, 1995;
Gilbert, 1997). Gilbert (1997) pointed out
that the series of articles demonstrating differences in the conserved phylotypic stage
of vertebrates (Richardson et al., 1997;
Richardson, 1999) was a backlash from the
focus on similarity. Clearly both similarities
and differences at all stages of development
are of interest to developmental biologists.
In contrast, phylogenetic systematists have
focused their attention first on determining
which similarities and differences may be
homologous and then on distinguishing the
primitive homologous features from derived. Shared derived features are then used
to reconstruct phylogenies, and from this,
the pattern of developmental character evolution hopefully emerges. Both developmental and systematic perspectives are necessary in developmental evolutionary biology, and each ‘‘reciprocally illuminates’’
(Hennig, 1966) the other.
The interests of systematists and developmental biologists converge on various
fundamental questions regarding the evolution of development such as ‘‘What are
the patterns of developmental evolution?’’;
‘‘How have evolutionary changes in developmental timing affected evolution?’’;
‘‘How do ontogenetic mechanisms generate, constrain or channel evolutionary variation?’’ As demonstrated in the vertebrate
limb example, questions of character association (Hox gene expression and skeletal
morphology) are of significant mutual interest.
1988; Wagner, 1989a, b; Hauser and
Presch, 1991). The empirical and theoretical assumptions underlying the use of developmental data in these ways are debated
in the systematic literature (Lundberg,
1973; De Queiroz, 1985; Mabee, 1989,
1993; reviewed in Mabee, 2000). The use
of development as a source of characters for
phylogenetic analysis (Hennig, 1966; Wiley, 1981; Alberch, 1985; De Queiroz,
1985; Mabee, 1993), however, is generally
viewed without contest and with great interest. Because most developmental biology
is done on model organisms (Kellogg and
Shaffer, 1993; Bolker, 1995) systematists
hoping to incorporate molecular developmental information into systematic studies
are limited to few species. In addition, few
of the complex morphological characters of
systematists can be explained in terms of
known genetic cascades (Holland, 1996),
and thus the developmental data at hand for
systematists consist mainly of descriptions
of morphological transformation. Gathering
even these data from the typical number of
taxa in a phylogenetic study is very timeintensive. In addition, incorporating the
transformational component of development into character hypotheses is not
straightforward, and further research and
experimentation into methods of character
coding are necessary (Mabee, 2000). Thus
in practice, very few developmental characters are used in phylogenetic analyses.
The specific uses of phylogenetic hypotheses in developmental biology are: 1)
A phylogeny can allow selection of most
appropriate taxa for study of particular
questions. 2) The evolutionary history of
developmental features can be traced. Ancestral (primitive) characters can be separated from derived, and the evolutionary sequence of changes in development can be
ascertained. Questions such as ‘‘At what
level is a developmental mechanism conserved?’’; ‘‘Where phylogenetically, did
this feature evolve?’’; ‘‘Where in the phylogeny was it modified?’’ can be addressed.
3) Predictions can be made regarding the
characters (e.g., type of developmental
mechanism) in unexamined taxa. 4) Multiple characters (e.g., morphological and developmental) can be mapped together on
DEVELOPMENT
AND
the phylogeny and patterns of character association can be quantified and tested. A
developmental basis of correlations found
might be hypothesized and tested. 5) Origins and levels of convergence and parallelism (homoplasy) can be quantified and
analyzed.
797
and phylogenetic tools for character optimization and correlation analysis needs to
be more prominent. The absence of critical
phylogenetic and morphological information shortchanges these devo/evo hypotheses. The reach of devo/evo must be extended to encompass the considerable comparative morphological data at hand and these
critical phylogenetic tools.
Conclusions
Robust hypotheses about the patterns and
processes of developmental evolution are
dependent upon the structure of phylogenetic hypotheses, the incorporation of morphological data, and the use of phylogenetic
methods for testing character associations.
By mapping homologous skeletal elements
of the vertebrate forelimb onto their phylogeny, an explicit alternative interpretation
of the significance of Hox gene expression
emerges. Since teleosts and tetrapods share
no homologous skeletal forelimb elements,
similarities and differences in Hox patterns
must be reinterpreted accordingly. Specifically, the presence of the phase III Hox pattern in tetrapods may not be correlated with
digits but rather may simply be the normal
expression pattern of a metapterygium.
More explicit, rigorous hypotheses can be
developed in devo/evo by using morphological data and phylogenetic trees and
methods.
ACKNOWLEDGMENTS
I thank R. Burian, S. Gilbert, B. Swalla
for their interest in incorporating systematics into this symposium. B. Swalla and two
anonymous reviewers provided helpful
comments and perspectives on this manuscript. Polyodon specimens were made
available by H. Bollig from Gavin’s Point
National Fish Hatchery. M. Noordsy, K.
Sykes, and P. Crotwell were helpful in obtaining images and editing this manuscript.
This work was supported in part by the
South Dakota National Science Foundation
Experimental Program to Stimulate Competitive Research (EPSCoR) and a National
Science Foundation grant (DEB-9408287)
to P.M.M.
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
Recommendations
A phylogenetic approach and morphological data are necessary for assembling
the most complete syntheses of development and evolution (Swalla and Collazo,
2000), yet phylogenetic techniques are increasingly sophisticated (Hillis et al.,
1996), and morphologists have accumulated a vast literature of comparative data.
Thus a ‘‘p.c.’’ (phylogenetically correct)
devo/evo study, though conceptually simple, may be difficult for a single researcher
to carry out. The steps in such a study involve: 1) choosing a monophyletic study
taxon (ingroup); 2) identifying (or generating) the best tree for this taxon and its
outgroups; 3) choosing the most appropriate taxa for developmental analysis within
the ingroup; 4) collecting comparative developmental data for selected taxa (to be
truly comparable these data must be homologous among taxa); 5) mapping the data
on the tree(s) chosen in step 2; 6) optimizing characters to determine ancestral states,
timing and direction of evolutionary
change; and 7) doing necessary tests for
character correlation (Maddison and Maddison, 1992; Cunningham et al., 1998; Cunningham, 1999). Collaboration with a specialist—a morphologist who knows phylogenetics or with both a phylogeneticist and
a morphologist—is perhaps the most reasonable solution to the potential complexity
involved in the above program of study.
Phylogenetics has been incorporated in
almost every subdiscipline in comparative
biology because of the need for evolutionary rigor in hypotheses of character evolution. Many comparative biologists, including comparative developmental biologists,
appreciate the importance of phylogenetic
methods for interpreting biological patterns
and processes. However, in the formulation
of devo/evo hypotheses, an explicit and
necessary role for phylogenetic hypotheses
PHYLOGENETICS
798
PAULA M. MABEE
REFERENCES
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
Ahlberg, P. E. 1992. Coelacanth fins and evolution.
Nature 358:459.
Ahlberg, P. E. and A. R. Milner. 1994. The origin and
early diversification of tetrapods. Nature 368:507–
514.
Alberch, P. 1985. Problems with the interpretation of
developmental sequences. Syst. Zool. 34:46–58.
Arthur, W. 1997. The origin of animal body plans.
Cambridge University Press, Melbourne, Australia.
Arthur, W., T. Jowett, and A. Panchen. 1999. Segments, limbs, homology, and co-option. Evolution
and Development 1:74–76.
Bolker, J. A. 1995. Model systems in developmental
biology. BioEssays 17:451–455.
Brooks, D. R. and D. A. McLennan. 1991. Phylogeny,
Ecology, and Behavior. University of Chicago
Press, Chicago.
Carroll, R. L. 1997. Development and evolution. In
Patterns and processes of vertebrate evolution,
pp. 212–265. Cambridge University Press, New
York.
Coates, M. I. 1994. The origin of vertebrate limbs.
Development Supplement: 169–180.
Coates, M. I. and J. A. Clack. 1990. Polydactyly in
the earliest known tetrapod limbs. Nature 347:66–
69.
Coates, M. I. and M. J. Cohn. 1998. Fins, limbs, and
tails: Outgrowths and axial patterning in vertebrate evolution. BioEssays 20:371–381.
Compagno, L. J. 1973. Interrelationships of living
elasmobranchs. Zool. J. Linn. Soc. 53:15–61.
Cunningham, C. W. 1999. Some limitations of ancestral character-state reconstruction when testing
evolutionary hypotheses. Syst. Biol. 48:665–674.
Cunningham, C. W., K. E. Omland, and T. H. Oakley.
1998. Reconstructing ancestral character states: A
critical appraisal. TREE 13:361–366.
de Carvalho, M. R. 1996. Higher-level elasmobranch
phylogeny, basal squaleans, and paraphyly. In M.
L. J. Stiassny, L. R. Parenti, and G. D. Johnson
(eds.), Interrelationships of fishes, pp. 35–62. Academic Press, San Diego.
De Queiroz, K. 1985. The ontogenetic method for determining character polarity and its relevance to
phylogenetic systematics. Syst. Zool. 34:280–299.
Donoghue, M. L., R. H. Ree, and D. A. Baum. 1998.
Phylogeny and the evolution of flower symmetry
in the Asteridae. Trends Plant Sci. 3:311–317.
Farris, J. S. 1970. Methods for computing Wagner
Trees. Syst. Zool. 19:83–92.
Farris, J. S. 1983. The logical basis of phylogenetic
analysis. In N. I. Platnick and V. A. Funk (eds.),
Advances in cladistics, Vol. 2, pp. 7–36. Columbia
University Press, New York.
Fitch, W. M. 1971. Toward defining the course of evolution: Minimum change for a specific tree topology. Syst. Zool. 20:406–416.
Forey, P. L. and P. Janvier. 1993. Agnathans and the
origin of jawed vertebrates. Nature 361:129–134.
Frumhoff, P. C. and H. K. Reeve. 1994. Using phylogenies to test hypotheses of adaptation: A critique
of some current proposals. Evolution 48:172–180.
Gardiner, B. G. 1984. The relationships of the palaeoniscid fishes, a review based on new specimens of
Mimia and Moythomasia from the Upper Devonian of Western Australia. Bull. Br. Mus. Nat.
Hist. Geol. 37:173–428.
Gerard, M., D. Duboule, and J. C. Zakany. 1993. Cooperation of regulatory elements involved in the
activation of the Hoxd-11 gene. Compt. R. Acad.
Sci. III 316:985–994.
Gilbert, S. F. 1997. http://zygote.swarthmore.edu/
evo5.html.
Goodrich, E. S. 1930. Studies on the structure and
development of vertebrates. Macmillan and Company, London.
Grandel, H. and S. Schulte-Merker. 1998. The development of the paired fins in the zebrafish (Danio
rerio). Mechanisms of Development 79:99–120.
Hadfield, K. A., B. J. Swalla, and W. R. Jeffery. 1995.
Multiple origins of anural development in ascidians inferred from rDNA sequences. J. Mol. Evol.
40:413–427.
Harvey, P. H. and M. D. Pagel. 1991. The comparative
method in evolutionary biology. Oxford University Press, Oxford.
Hauser, D. and W. Presch. 1991. The effects of ordered
characters on phylogenetic reconstruction. Cladistics 7:243–265.
Hennig, W. 1966. Phylogenetic systematics. University
of Illinois Press, Urbana.
Hillis, D. M., C. Moritz, and B. K. Mable. 1996. Molecular systematics. Sinauer, Sunderland, Massachusetts.
Holland, P. W. H. 1996. Molecular biology of lancelets: Insights into development and evolution. Isr.
J. Zool. 42:S247–S272.
Holland, P. W. H. and J. Garcia-Fernandez. 1996. Hox
genes and chordate evolution. Dev. Biol. 173:
382–395.
Huey, R. B. and A. F. Bennett. 1986. A comparative
approach to field and laboratory studies in evolutionary biology. In M. E. Feder and G. V. Lauder (eds.), Predator–prey relationships: Perspectives and approaches from the study of lower vertebrates, pp. 82–98. University Chicago Press,
Chicago, Illinois.
Hull, D. L. 1988. Science as a process. University of
Chicago Press, Chicago.
Janvier, P. 1981. The phylogeny of the Craniata, with
particular reference to the significance of fossil
‘Agnathans.’ J. Vertebr. Paleontol. 1:121–159.
Janvier, P. 1999. Catching the first fish. Nature 402:
21–22.
Jarvik, E. 1980. Basic structure and evolution of vertebrates. Academic Press, London.
Jessen, H. L. 1972. Schultergurtel und Pectoralflosse
bei Actinopterygiern. Fossils Strata 1:1–101.
Johnson, G. D. and C. Patterson. 1996. Relationships
of lower euteleostean fishes. In M. L. J. Stiassny,
L. R. Parenti, and G. D. Johnson (eds.), Interrelationships of fishes, pp. 251–332. Academic
Press, San Diego.
Kellogg, E. A. and H. B. Shaffer. 1993. Model organisms in evolutionary studies. Syst. Biol. 42:409–
414.
DEVELOPMENT
AND
799
lution and development. Anat. Embryol. 196:91–
106.
Rosen, D. E., P. L. Forey, B. G. Gardiner, and C. Patterson. 1981. Lungfishes, tetrapods, paleontology,
and plesiomorphy. Bull. Am. Mus. Nat. Hist. 167:
159–276.
Roth, V. L. 1984. On homology. Biol. J. Linn. Soc.
22:13–29.
Roth, V. L. 1988. The biological basis of homology.
In C. J. Humphries (ed.), Ontogeny and systematics, pp. 1–26. Columbia University Press, New
York.
Ryan, M. J. 1996. Phylogenetics in behavior: Some
cautions and expectations. In E. P. Martins (ed.),
Phylogenies and the comparative method in animal behavior, pp. 1–21. Oxford University Press,
New York.
Sanderson, M. J. 1991. In search of homoplastic tendencies: Statistical inference of topological patterns in homoplasy. Evolution 45:351–358.
Schultz, T. R., R. B. Crocroft, and G. A. Churchill.
1996. The reconstruction of ancestral character
states. Evolution 50:504–511.
Shirai, S. 1996. Phylogenetic interrelationships of neoselachians (Chondrichthyes: Euselachii). In M. L.
J. Stiassny, L. R. Parenti and G. D. Johnson (eds.),
Interrelationships of fishes, pp. 9–34. Academic
Press, San Diego.
Shubin, N. H. and P. Alberch. 1986. A morphogenetic
approach to the origin and basic organization of
the tetrapod limb. Evol. Biol. 20:319–387.
Shubin, N., C. J. Tabin, and S. B. Carroll. 1997. Fossils, genes and the evolution of animal limbs. Nature 388:638–648.
Sordino, P., F. van der Hoeven, and D. Duboule. 1995.
Hox gene expression in teleost fins and the origin
of vertebrate digits. Nature 375:678–681.
Sordino, P. and D. Duboule. 1996. A molecular approach to the evolution of vertebrate paired appendages. TREE 11:114–119.
Swalla, B. J. and A. Collazo. 2000. Systematics and
the evolution of developmental patterns. Syst.
Biol. 49:1–2.
Swofford, D. L. and W. P. Maddison. 1992. Parsimony,
character-state reconstructions, and evolutionary
inferences. In R. L. Mayden (ed.), Systematics,
historical ecology, and North American freshwater fishes, pp. 186–223. Stanford University Press,
Stanford.
Tabin, C. J., S. B. Carroll, and G. Panganiban. 1999.
Out on a limb: Parallels in vertebrate and invertebrate limb patterning and the origin of appendages. Amer. Zool. 39:650–663.
van der Hoeven, F., J. Zakany, and D. Duboule. 1996.
Gene transpositions in the HoxD complex reveal
a hierarchy of regulatory controls. Cell 85:1025–
1035.
Wada, H., H. Saiga, N. Satoh, and P. W. H. Holland.
1998. Tripartite organization of the ancestral chordate brain and the antiquity of placodes: Insights
from ascidian Pax-2/5/8, Hox and Otx genes. Development 125:1113–1122.
Wagner, G. P. 1989a. The biological homology concept. Annu. Rev. Ecol. Syst. 20:51–69.
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014
Knoll, A. H. and S. B. Carroll. 1999. Early animal
evolution: Emerging views from comparative biology and geology. Science 284:2129–2137.
Lauder, G. V. and K. F. Liem. 1983. The evolution and
interrelationships of the actinopterygian fishes.
Bull. Mus. Comp. Zool. 150:95–197.
Lundberg, J. G. 1973. More on primitiveness, higher
level phylogenies and ontogenetic transformations. Syst. Zool. 22:327–329.
Mabee, P. M. 1989. An empirical rejection of the ontogenetic polarity criterion. Cladistics 5:409–416.
Mabee, P. M. 1993. Phylogenetic interpretation of ontogenetic change: Sorting out the actual and artefactual in an empirical case study of centrarchid
fishes. Zool. J. Linn. Soc. 107:175–291.
Mabee, P. M. 2000. The usefulness of ontogeny in interpreting morphological characters. In J. Wiens
(ed.), Phylogenetic analysis of morphological
data, pp. 84–114. Smithsonian Press, Washington.
Maddison, D. R. 1994. Phylogenetic methods for inferring the evolutionary history and processes of
change in discretely valued characters. Annu. Rev.
Ecol. Syst. 39:267–292.
Maddison, W. P. 1990. A method for testing the correlated evolution of two binary characters: Are
gains or losses concentrated on certain branches
of a phylogenetic tree? Evolution 44:539–557.
Maddison, W. P. and D. R. Maddison. 1992. MacClade: Analysis of phylogeny and character evolution, version 3.0. Sinauer, Sunderland, Massachusetts.
Maisey, J. G. 1984. Chondrichthyan phylogeny: A
look at the evidence. J. Vertebr. Paleontol. 4:359–
371.
Maisey, J. G. 1986. Heads and tails: A chordate phylogeny. Cladistics 2:201–256.
Nelson, C. E. and C. Tabin. 1995. Footnote on limb
evolution. Nature 375:630–631.
Nelson, C. E., B. A. Morgan, A. C. Burke, E. Laufer,
E. DiMambro, L. C. Murtaugh, E. Gonzales, L.
Tessarollo, L. Parada, and C. Tabin. 1996. Analysis of Hox gene expression in the chick limb bud.
Development 122:1449–1466.
Nelson, G. J. 1973. The higher-level phylogeny of vertebrates. Syst. Zool. 22:87–91.
Nelson, G. J. 1978. Ontogeny, phylogeny, paleontology, and the biogenetic law. Syst. Zool. 27:324–
345.
Omland, K. E. 1999. Assumptions and challenges of
ancestral state reconstructions. Syst. Biol. 48:604–
611.
Patterson, C. 1982. Morphological characters and homology. In K. A. Joysey and A. E. Friday (eds.),
Problems of phylogenetic reconstruction, Vol. 21,
pp. 21–74. Academic Press, London.
Raff, R. A. 1996. The shape of life. University of Chicago Press, Chicago.
Richardson, M. K. 1999. Vertebrate evolution: The developmental origins of adult variation. BioEssays
21:604–613.
Richardson, M. K., J. Hanken, M. L. Gooneratne, C.
Pieau, A. Raynaud, and L. Selwood. 1997. There
is no highly conserved embryonic stage in the vertebrates: Implications for current theories of evo-
PHYLOGENETICS
800
PAULA M. MABEE
Wagner, G. P. 1989b. The origin of morphological
characters and the biological basis of homology.
Evolution 43:1157–1171.
Wake, D. B., P. M. Mabee, J. Hanken, and G. Wagner.
1991. Development and evolution—the emergence of a new field. In E. C. Dudley (ed.), Proceedings of the Fourth International Congress of
Systematic and Evolutionary Biology, ICSEB IV,
1, pp. 582–588. Dioscorides, Portland.
Wiley, E. O. 1981. Phylogenetics: The theory and
practice of phylogenetic systematics. John Wiley,
New York.
Wray, G. A. 1996. Parallel evolution of nonfeeding
larvae in echinoids. Syst. Biol. 45:308–322.
Downloaded from http://icb.oxfordjournals.org/ at Pennsylvania State University on March 5, 2014