DEVELOPMENTAL BIOLOGY 173, 357372 (1996)

ARTICLE NO. 0032

REVIEW
Resynthesizing Evolutionary
and Developmental Biology
Scott F. Gilbert,*,1 John M. Opitz, and Rudolf A. Raff
*Department of Biology, Swarthmore College, Swarthmore, Pennsylvania 19081; Foundation
for Developmental and Medical Genetics, FRB-Suite 229, 100 Neill Avenue, Helena, Montana
59601 and Department of History, Montana State University, Bozeman, Montana 59715; and
Department of Biology, and Institute of Molecular and Cellular Biology,
Jordan Hall, Indiana University, Bloomington, Indiana 47405

A new and more robust evolutionary synthesis is emerging that attempts to explain macroevolution as well as microevolutionary events. This new synthesis emphasizes three morphological areas of biology that had been marginalized by the
Modern Synthesis of genetics and evolution: embryology, macroevolution, and homology. The foundations for this new
synthesis have been provided by new findings from developmental genetics and from the reinterpretation of the fossil
record. In this nascent synthesis, macroevolutionary questions are not seen as being soluble by population genetics, and
the developmental actions of genes involved with growth and cell specification are seen as being critical for the formation
of higher taxa. In addition to discovering the remarkable homologies of homeobox genes and their domains of expression,
developmental genetics has recently proposed homologies of process that supplement the older homologies of structure.
Homologous developmental pathways, such those involving the wnt genes, are seen in numerous embryonic processes,
and they are seen occurring in discrete regions, the morphogenetic fields. These fields (which exemplify the modular nature
of developing embryos) are proposed to mediate between genotype and phenotype. Just as the cell (and not its genome)
functions as the unit of organic structure and function, so the morphogenetic field (and not the genes or the cells) is seen
as a major unit of ontogeny whose changes bring about changes in evolution. q 1996 Academic Press, Inc.

INTRODUCTION: THE GENETIC

REDEFINITION OF EVOLUTION AND
THE ECLIPSE OF MACROEVOLUTION
AND HOMOLOGY
In 1932, Thomas Hunt Morgan published his famous address on The Rise of Genetics. Delivered originally at the
Sixth International Congress of Genetics at Cornell University, this would become the historical statement of the field
by its acknowledged leader. It would also become the model
for nearly all subsequent histories of genetics, many of them
written by those tracing their lineage to the Fly Room. It
came at a defining moment for the rapidly growing field of
genetics. Although initially a unified science, genetics and
1
To whom correspondence should be addressed. Fax: 610-3288663.

embryology diverged from each other during the 1920s, and

by the 1930s, genetics and embryology had their own rules
of evidence, their own paradigmatic experiments, their own
favored organisms, their own professors, their own journals,
and most importantly, their own vocabularies (Allen, 1978;
Gilbert, 1978, 1988). In 1926, Morgan had formally separated genetics from embryology. Now he would go further,
proclaiming that genetics had superseded embryology and
had put order into the study of evolution. Morgan (1932a)
contrasted the genetic approach to evolution with that of
the old school of morphology and comparative anatomy.
He claimed that genetics has made a very important contribution to evolution, especially when it is recalled that it
has brought to the subject an exact scientific method of
procedure. That same year (Morgan, 1932b), he would contend that genetic studies furnish us today with ideas for
an objective study of evolution in striking contrast to the
older speculative method of treating evolution as a problem

0012-1606/96 $12.00
Copyright q 1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

357

dba

Dev Bio

358

Gilbert, Opitz, and Raff

of history. No wonder paleontologists such as W. K. Gregory (1917) had written about genetics versus paleontology: Morgan believed that Genetics brought evolutionary
biology out of natural history into the domain of science.
In 1937, Morgans student, Theodosius Dobzhansky, carried this idea further and took the bold step of redefining
evolution as changes in gene frequency. Instead of being a
phenotypic science analyzing changes in fossil morphology,
embryonic structures, or the alterations that make a structure adaptive in a particular environment, evolution became the epiphenomenon of the genetics of populations.
The changes in gene frequency inferred by melanotic moth
wings or beetle elytra could model how fish gave rise to
amphibians. The Modern Synthesis supported population
genetics as the major focus of evolutionary science and
viewed genetics as Darwins missing evidence (Kettlewell, 1959). Thus, evolution could be competely explained by the mutation and separation of genes. Numerous
biologists, especially paleontologists and the Soviet school
of population biology, had argued against this view. I. A.
Filipchenko (1929) coined the terms microevolution and
macroevolution and argued that one could not be inferred
from the other. Microevolution concerned the origin of varieties and races within species. Macroevolution concerned
the origins of higher taxa. Originally, H. F. Osborn (1925),
G. G. Simpson, and other American paleontologists did not
accept the view that the fossil record could be explained by
the accumulation of minute selectable changes over millions of years. But eventually, the Soviet school of population genetics was liquidated, and the American paleontologists retreated into their museums (Adams, 1990). Population genetics became the predominant explanatory mode
for evolutionary biology, and by 1951, Dobzhansky could
confidently declare, Evolution is a change in the genetic
composition of populations. The study of mechanisms of
evolution falls within the province of population genetics.
Thus, evolution was seen as a subset of the formal mathematics of population genetics (see Gottlieb, 1992), and there
was nothing in evolutionary biology that fell outside of it.
One of the major tenets of the Modern Synthesis has been
that of extrapolation: the phenomena of macroevolution,
the evolution of species and higher taxa, are fully explained
by the microevolutionary processes that gives rise to varieties within species. Macroevolution can be reduced to microevolution. That is, the origins of higher taxa can be explained by population genetics.
There were several reasons for the success of the population genetic approach to evolution. First and foremost, it
got results. One could not expect to see species or phyletic
change over a lifetime, but microevolutionary changes
could be observed in the field or in the laboratory. Moreover,
unlike most of biology, these results were phrased in the
unambiguous language of mathematics. There were also
social factors that hastened the hegemony of genetic approaches to evolution over any other. First, the population
genetic approach to evolution was readily funded by the
Atomic Energy Commission. Whereas most evolutionary

studies had difficulty getting funds and students, concerns

about the genetic effects of radiation enabled Dobzhansky
and others a constant supply of money and graduate students (Beatty, 1994). Second, the linkage of evolution and
genetics fit into certain social agendas. As Paul (1988) has
shown, Dobzhansky and others viewed the population genetic model of adaptation as undermining the racial and
class associations of fitness. Moreover, there was the
threat of Creationism. In the United States, evolution is still
so suspect that no National Science Foundation program is
designated as Evolutionary Biology. In the 1930s and
1940s, it was even more suspect. Genetics, however, was
(and is) seen as being true and economically important. If
evolution were merely a change in the genetic composition
of a population, then evolution is a mathematically proven
fact. Evolution is nought but genetics writ large. In the Soviet Union, the same phenomenon occurred in reverse. Official ideology held Darwinism in enormous respect, but
Genetics was a suspect bourgois science. By identifying genetics with Darwinism, genetics was allowed to operate (at
least for a time) in the Soviet Union (Adams, 1990).
Genetics also provided a mechanism for evolution when
no other mechanism was available. If there were a Modern
Synthesis between genetics and evolution, there had to
have been some Unmodern Synthesis that it replaced.
This Unmodern Synthesis was the notion that evolution
was caused by changes in development. The syntheses of
E. Haeckel, E. Metchnikoff, A. Weismann, W. K. Brooks,
and others were that of evolution and embryology. Haeckels Biogenetic Law had superseded all the other developmental syntheses, and by the 1930s, this synthesis had become both racist and scientifically untenable (see Gasman,
1971; Gould, 1977). It was an easy target for both geneticists
and embryologists (such as W. Garstang and N. J. Berrill) to
destroy. But in the 1930s and 1940s, embryology had nothing new to substitute for this discredited notion. In fact,
embryologists were no longer interested in evolution and
had separated themselves from evolutionary biology in an
attempt to become more scientific, i.e., experimental (Allen, 1978; Maienschein, 1991). Genetics readily filled this
vacuum, and the Modern Synthesis substituted genetics for
embryology as the motor for evolution. Thus, embryologywhich had previously been the handmaid to evolution (Baldwin, 1902) and which Darwin perceived as his
major source of evidencegave way to genetics.
One obvious and immediate casualty of this replacement
was the autonomy of macroevolution. Macroevolution was
completely explainable by the processes of microevolution.
It had no status of its own. Another casualty of the population genetical approach to evolution was the notion of homology. Homology was popularized by Darwins major adversary, Richard Owen (1849), who saw homologous structures as representing the same organ in all its variety of
forms and functions. It thereby related organisms to one
another by particular affinities of structure. The arms and
legs in humans were not only serially homologous to each
other within the organism, but also specially homologous

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

359

A New Evolutionary Synthesis

to the fore- and hind-limbs of Beasts, the wings and legs

in Bats and Birds, and the pectoral fins and ventral fins
of fishes. (Evolutionary biologists would now call these
structures historically homologous or orthologous.) Indeed,
the general homologous plan of all vertebrates could be
discerned by anatomical studies. Thomas Huxley (1858)
emphasized that these homologies were often seen more
clearly during developmental stages of these organisms, and
Charles Darwin used homologies to indicate common descent (as opposed to Owens view that they indicated construction on the same rational plan). However, homologies
merely offered evidence for the operation of evolution. They
did not provide a mechanism for evolution. Natural selection and sexual selection, the two mechanisms favored by
Darwin, were both based on adaptations in organisms
within a species competing for reproductive success. Competition would create new forms out of old ones. Evolution
depended upon intraspecies differences between organisms,
not interspecies similarities. And genes manifested themselves as differences. Homologyand the construction of
phylogenetic trees based on common embryonic structuresseemed old-fashioned and unscientific compared to
the mathematical elegance of population genetics.
Indeed, even before the rise of genetics, studies of embryonic homologies were going against the grain of the new
evolutionary biology. This is clearly seen in the Marine
Biology Laboratory Lectures of 1898. One of the speakers,
embryologist E. B. Wilson, delivered a classic paper on Cell
lineage and ancestral reminiscences, demonstrating that
the cleavage of flatworms, molluscs, and annelids all shared
a homologous pattern. Thus, a gap that seemed hopelessly wide was finally bridged. He was followed by an
equally famous embryologist, F. R. Lillie, who also spoke on
molluscan cleavage. However, Lillie discussed deviations in
embryonic development which produced selectable adaptations. He argued that modern evolutionary biology would
do better to concentrate on changes that enabled organisms
to survive in particular environments than to focus on ancestral homologies that united animals into lines of descent.
Homology was moving into the background.

THE GENETIC REDEFINING OF

EXPERIMENTAL EMBRYOLOGY: THE
ECLIPSE OF THE MORPHOGENETIC FIELD
If evolution became an epiphenomenon or subset of genetics, then a similar change happened to embryology. The
story of the dismissal of embryology from the Modern Synthesis has been repeated many times (e.g., Hamburger, 1988;
Gottlieb, 1992), but the reasons for this removal remain
obscure. We will try to show here that there are several
reasons why embryology could not fit into the synthesis
and one of them was that its main explanatory entity, the
morphogenetic field, was viewed as a threat to the gene as
the unit of ontogeny and phylogeny. From the 1920s

through the middle of the 1930s, embryology experienced

a Renaissance (see Oppenheimer, 1966). This was the age
of Spemanns laboratory and the foundations of the Organizer; it was the age of Harrisons demonstration of limb
polarity and of Hamburgers and Weiss studies on neuron
growth and specificity; it was the time of Horstadius and
Childs gradients, Willier and Rawles demonstration of the
neural crest cell migrations, and Witschis observations of
sex determination and gonad differentiation. Needham,
Waddington, and Brachet were constructing a biochemical
embryology, and it appeared as if the basis of morphogenesis
was going to be discovered. The research program of this
optimistic and robust embryology was Gestaltungsgesetze,
the attempt to discover the laws of ordered form (Needham,
1931). The basic paradigm of embryology, the idea that gave
it structure and coherence, was the morphogenetic field.
It is difficult to realize how powerful the concept of the
morphogenetic field used to be. It was one of those notions
that was so powerful as to be assumed rather than continually proven (Oppenheimer, 1966). To Needham (1950), the
field gave powerful aid to the codification of Gestaltungsgesetze. . . . The concept of morphogenetic fields
within the embryo was postulated by Boveri (1910; see
Sander, 1994) and given explicit definition by Alexander
Gurwitsch (1910, 1912, 1922), who initially called them
Geschehnsfeld and Kraftfeld, and finally (1922) Embryonales Feld. This idea was popularized through the limb
transplantation experiments of Harrison (1918; see Haraway, 1976). Harrison demonstrated that the newt neurula
contained two discs of cells which could form a forelimb
when transplanted to another region of the embryo. Moreover, cells within this field could regulate. If a limb field
were cut in half and the two halves transplanted to different
locations, each half would form a complete limb. Conversely, if two half-limbs were grafted together in the same
orientation, the fields could regulate to form one normal
limb. If undetermined cells or tissues were introduced into
the field domain, they became organized and incorporated
into the limb. Harrison (echoing Driesch) called this a selfdifferentiating equipotential system. Harrisons friend,
Hans Spemann (1921), reinvented this concept as an Organisationsfeld and said that the dorsal blastopore lip established such a field of organization. Paul Weiss (1923)
would come to similar concepts and names (perhaps independently) and he would give this concept an important
theoretical basis. These fields designated areas of embryological information, bound by physical substrates. The components of these fields created a web of interactions such
that any cell was defined by its position within its respective field.
The morphogenetic fieldlike the terms homology or
genemeant somewhat different things to different people.
This might be expected when the term is applied to systems
as diverse as regenerating planaria, neural induction, and
limb determination (see Herrmann, 1964). Like an electromagnetic field, the term denoted both informational and
regional relationships. Needham (1950) approved of the use

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

360

Gilbert, Opitz, and Raff

of fields to explain embryonic phenomena, and he combined

the views of Spemann, Waddington, and Weiss in the following definition:
A morphogenetic field is a system of order such that the positions
taken up by unstable entities in one part of the system bear a
definite relation to the position taken up by other unstable entites
in other parts of the system. The field effect is constituted by their
several equilibrium positions. A field is bound to a particular substratum from which a dynamic pattern arises. It is heteroaxial and
heteropolar, has recognizably distinct districts, and can, like a magnetic field, maintain its pattern when its mass is either reduced or
increased. It can fuse with a similar pattern entering with new
material if the axial orientation is favorable. The morphogenetic
gradient is a special limited case of the morphogenetic field.

Paul Weiss textbook Principles of Development (1939)

popularized the field concept and used it as an organizing
principle for all of embryology. Weiss noted that the field
concept has been extensively adopted by embryologists,
and he set out to provide some structure to this flexible
concept. His concept of the field was based on purely empirical evidence, and he concluded that the field had the attributes of individuality, heteropolarity, and gradation. Moreover, not only did most developmental phenomena show
these field properties, but the field had a real, physical, existence. The field concept is not only a useful circumlocution, but an expression of physical reality. This elevated
the field to the dignity of an object of research, and it
imposed a duty to study it just as one would study any
newly discovered natural phenomenon. If the term field
were mistaken for a sort of narcotic devised to appease the
mental discomfort arising from our profound ignorance of
the problem of organization, its use would be highly inexpedient.
In addition to Weiss highly interactive, ecosystem-like
fields existed a related model, the gradient-field. This was
the brainchild of Gavin de Beer and was popularized in Huxley and de Beers Elements of Experimental Embryology
(1934). Such a gradient-field would combine the morphogenetic field concept with the gradient concept. As De Robertis and co-workers (1991) have noted, this concept had
three sources of evidence. First, the was the Gefall (gradient)
hypothesis of Boveri, whereby differential concentrations
of substances could determine cell fate. Second, there were
the experiments of Swett (1923) which showed that the
maximum forelimb-forming ability is found in the anterodorsal region of the forelimb field and decreases gradually
from there to the rest of the field. Third, experiments on
regenerating planaria showed that whether a particular
group of cells regenerated a head or a tail depended solely
on the cells attached to it. If the cells were at the anterior
tip of the amputation, they became head; if they were at
the posterior end of the amputation, they became tail. Moreover, if both head and tail were cut off the planarian, whichever cells were anterior formed head; whichever were posterior formed tail. The field-like nature of this phenomenon

was shown by making deep cuts into the head region. If

prevented from re-fusing, each portion would form a new
and complete head. Child (1915, 1941) showed that there
was an axial gradient to this regeneration potential. The
percentage of animals able to regenerate heads decreased as
the distance of the amputation site from the anterior. (Weiss
criticized the linkage of fields with gradients. The gradient,
he felt, was just a symbol to indicate the direction and
rapidity of the decline in field activity.) There were several
related, but competing, notions of what exactly a field was.
Then, as Opitz (1985) remarked,
In one of the most astounding developments in Western scientific
history, the gradient-field, or epimorphic field concept, as embodied
in normal ontogeny and as studied by experimental embryologists,
seems to have simply vanished from the intellectual patrimony of
Western biologists.

What destroyed the morphogenetic field? One answer is

that nothing destroyed the morphogenetic field. No data
were presented arguing that the idea was wrong or that
fields did not exist. Rather, the morphogenetic field was
eclipsed and ignored. There were several reasons for this
eclipse. First, biochemical techniques were not good enough
to enable embryologists to examine field phenomena such
as limb polarity, neural tube patterning, and so forth. Second, there was the decline of funding for biological sciences
in Europe, especially in Germany, which had been the intellectual and institutional base of embryology. Third, there
was the rise of genetics with its alternative program for
development. This last point is critical, for just as evolution
became redefined as the study of changes in gene frequency,
so embryology became redefined as the science studying
changes in gene expression (Morgan, 1934). Since morphogenesis was subsumed in the larger category of gene expression, fields were not needed. Eventually, embryogenesis became synonymous with cell differentiation, and by 1948,
Sol Spiegelman could argue that cell differentiation was
synonymous with differential protein synthesis and could
be studied more readily in Escherichia coli or yeast than in
metazoan embryos. The formation of complex organs could
be seen as being caused by small changes in the gene expression, just as the evolutionary alterations of complex morphology could be effected by the accumulation of small gene
changes. Thus, two phenotypic sciences, embryology and
evolution, were given new, genotypic, definitions (see Gilbert, 1996a).
The Genetics program of biology was in direct opposition
to the concept of morphogenetic fields. Morgan, who had
once been second only to C. M. Child in his publication
record on gradient fields, blocked the attempts of Child and
his students to publish their findings. Morgan considered
such work old-fashioned and not good science (Mittman and
Fausto-Sterling, 1992). Indeed, Mitman and Fausto-Sterling
(1992) conclude that Morgan was so adamant about ridiculing the field notion because in the 1930s, the morphogenetic
field was an alternative to the gene as the unit of ontogeny.

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

361

A New Evolutionary Synthesis

Neither field nor gene had been seen. Both were postulated
on the results of experimental data. Both sought to explain
inheritance. In planaria, the inherited information could be
seen in the gradient which enabled the organism to form a
head at one end and a tail at the other. Upon splitting, each
half inherited the ability to make a whole and properly
organized animal. In Drosophila, several generations of flies
could inherit a trait according to strict statistical laws, suggesting the involvement of nuclear chromosomes. The gene
and the field were in opposition.
The geneticists and the embryologists ridiculed each others theories. In his aptly titled The Rise of Genetics,
Morgan laments,
If another branch of zoology that was actively cultivated at the end
of the last century had realized its ambitions, it might have been
possible to-day to bridge the gap between gene and character, but
despite its high-sounding name of Entwicklungsmechanik nothing
that was really quantitative or mechanistic was forthcoming. Instead, philosophical platitudes were invoked rather than experimentally determined factors. Then, too, experimental embryology
ran for a while after false gods that landed it in a maze of metaphysical subtleties.

Geneticists portrayed embryologists were seen as being oldfashioned, mystical, and metaphysical, enemies to good science. Because of these characteristics, they had failed to
achieve their goal of linking genes and characters. But this
was Morgans rhetoric; it was never the goal of most embryologists. Embryologists (as R. Goldschmidt noted in 1940)
had not been interested in gene expression; they had other
problems (induction and morphogenetic fields in the program of Gestaltungsgesetze) to keep themselves occupied.
Morgan presented no evidence against fields or gradients
(see Gilbert, manuscript submitted for publication). Rather,
these concepts were viewed as being mystical, holistic, relics of the past, not to be taken seriously in the new genebased reductionist biology.
Embryologists, on the other hand, saw genetics as no
more intellectual than...a game of cards. Certainly, most
embryologists did not feel that they needed to take genes
seriously. Embryologist N. J. Berrill (1941) said that he felt
that genes were statistically significant little devils collectively equivalent to one entelechy. Genes are not mentioned in most of the contemporary embryology texts (including Spemann, 1938), and Harrison (1937) could ask how
the geneticists could possibly say that genes controlled development when they could not explain how identical genes
in each cell created different cell types and when they could
not point to any examples of genes being active in early
development. Genes could determine the number of bristles
on a flys back, but they could not determine how a fly
constructed its back in the first place. The construction of
the organism was accomplished by fields. The contempt of
embryologists for evolutionary biology also helped write
embryology out of the synthesis (see Smocovitis, 1994).
They considered embryology as an experimental discipline,

superior to the collecting and describing that characterized

genetics and evolution.
De Robertis and colleagues (1991) have suggested that
morphogenetic fields disappeared from the literature because they were abstract, almost metaphysical, conceits
that could only be revealed experimentally. However, at
the time, morphogenetic fields were no more abstract than
genes, and even geneticists such as Bateson and Goldschmidt admitted that the gene was a metaphysical concept
whose physical reality remained in doubt. Oppenheimer
(1966) suggests that the field concept died out because its
validity was taken so much for granted that nobody set
down to prove it. However, we would contend that morphogenetic fields disappeared from the literature because the
techniques to analyze them had not yet appeared and because they were eclipsed by the genetic explanation of development in which fields were not needed.
By the late 1930s, evidence was obtained for genetically
controlled programs of embryogenesis (Morgan, 1934;
Schultz, 1935; Beadle and Ephrussi, 1937), and mutations
were found that involved the early stages of animal development (Gluecksohn-Schoenheimer, 1938). The eclipse of the
field by the gene had been started. The success of the genetic
program is manifest in our being so ignorant of the power
that morphogenetic fields had prior to World War II and the
rise of Genetics.

THREE RE-DISCOVERIES
The Modern Synthesis is a remarkable achievement.
However, starting in the 1970s, many biologists began questioning its adequacy in explaining evolution. Genetics
might be adequate for explaining microevolution, but microevolutionary changes in gene frequency were not seen
as able to turn a reptile into a mammal or to convert a
fish into an amphibian. Microevolution looks at adaptations
that concern only the survival of the fittest, not the arrival
of the fittest. As Goodwin (1995) points out, the origin
of speciesDarwins problemremains unsolved. This
reexamining of the Modern Synthesis has led to three great
re-discoveries in modern biology. These are the simultaneous rediscoveries of macroevolution, homology, and the
morphogenetic field. A new synthesis is emerging from
these three areas, and this developmentally oriented synthesis may soon be able to explain macroevolutionary as well
as microevolutionary processes. The first condition for their
rediscovery came from scientists such as R. B. Goldschmidt
and C. H. Waddington, who saw that all changes important
in evolution are alterations in development. When we say
that the one-toed horse is derived from a five-toed ancestor,
we are saying that changes have occurred in the development of the limb cartilage cells. Some genes involved in
chondrocyte growth, placement, or differentiation have
changed. Evolution, to use Goldschmidts (1940) phrase, involves heritable changes of development. This can be represented as follows (Gilbert and Faber, 1996):

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

362

Gilbert, Opitz, and Raff

Functional Biology anatomy, physiology,

cell biology, gene expression
Developmental Biology d [functional biology]/dt
Evolutionary Biology d [developmental biology]/dt
Here, a tertium quid, development, has been imposed between Ernst Mayrs two categories of functional and evolutionary biology. This interpositioning is both conceptual
and physical. First, it suggests that to go from functional
biology to evolutionary biology without considering developmental biology is like going from displacement to acceleration without considering velocity. Second, it positions development as hierarchically between the two other categories and mediating between them. Development not only
is the agent through which these changes are effected, but
development constrains selection in its ability to produce
new phenotypes (Alberch, 1982). Third, it suggests that
there might be a physical substrate which accomplishes this
mediation. We suggest that the morphogenetic field is such
a substrate.

The Rediscovery of Macroevolution

The concept that macroevolution could not be derived
from microevolution remained as an underground current
in evolutionary theory. Every so often, it was brought to the
surface by developmentally oriented evolutionary biologists
such as Goldschmidt, Waddington, or de Beer. In 1940,
Richard Goldschmidt stated the challenge to those who proposed the Modern Synthesis. How could the origin of such
things as mammalian hair, aortic arches, mollusc shells,
cnidocysts, or the compound eye be explained by accumulation and selection of small mutants? But these attempts
to decouple microevolution from macroevolution were either ignored or marginalized (see Gilbert, 1994a).
Macroevolution was brought back as an autonomous entity only after Eldredge and Gould (1972), Stanley (1979),
and others postulated an alternative view to the gradualism
that characterized the Modern Synthesis. By 1980, Gould
claimed that the idea of gradual alleleic substitution as a
mode for all evolutionary change was effectively dead.
This view did not go unchallenged, and by 1982, Goulds
view had become more specific. It wasnt that the Modern
Synthesis was wrong; rather, it was incomplete. Nothing
about microevolutionary population genetics, or any other
aspect of microevolutionary theory, is wrong or inadequate
at its level. . . . But it is not everything (Gould, 1982; p.
104). While punctuated equilibrium remained a controversial theory, it did bring to light the question of the autonomy of macroevolution. Indeed, the failure of microevolutionary biology to distinguish between punctuated equlibrium and gradualism demonstrated its weakness when
applied to macroevolution (see Ayala, 1983). Molecular
studies (King and Wilson, 1975) were similarly pointing to
evolution at two levels, one molecular, the other morpho-

logical. Thus, by the early 1980s, numerous paleontologists

and evolutionary biologists (Gould, Stanley, Eldredge,
Verba, and most critically, Ayala) came to the conclusion
that although macroevolutionary phenomena were underlain by microevolutionary phenomena, the two areas were
autonomous and that macroevolutionary processes could
not be explained solely by microevolutionary events.

The Rediscovery of Homology

Homology is an important word again. Brian K. Hall
(1994) has recently published a 13-chapter discussion of its
meaning, and David Wake (1995) noted that whatever it
means, it is the most important concept in contemporary
biology:
Homology is the central concept for all biology. Whenever we say
that a mammalian hormone is the same as a fish hormone, that
a human sequence is the same as a sequence in a chimp or a
mouse, that a HOX gene is the same in a mouse, a fruit fly, a
frog, and a humaneven when we argue that discoveries about a
roundworm, a fruit fly, a frog, a mouse, or a chimp have relevance
to the human conditionwe have made a bold and direct statement about homology.

Homology was rediscovered almost simultaneously by

several groups of scientists, including molecular biologists,
developmental geneticists, clinical geneticists, and paleontologists. Paleontologists had continually been using the
term but dramatically reformulated it in the 1980s (Van
Valen, 1982; Roth, 1984; Wagner, 1984), largely as a result
of critiques of the adaptationist program in evolutionary
biology. Gould, in particular, hit vehemently against the
adaptationist paradigms of contemporary evolutionary biology and substituted a paradigm based on developmental
constraints and homology. In his paper with Richard Lewontin (1979), The Spandrels of San Marco, the idea of
developmental homology was reasserted, and the just-so
stories of the adaptationists were held up to ridicule.
Gould (1977) was able to go a step further than his predecessors by postulating mechanisms that would produce homologous structures and at the same time provide routes for
rapid morphological change. These mechanisms were heterochrony (changes in the relative timing of developmental
events) and allometry (differential growth of parts). Both
these mechanisms had been proposed earlier by developmentally oriented evolutionary biologists (heterochrony by
de Beer, 1940; allometry by Huxley, 1932), and Gould uses
them to demonstrate how developmental changes can rapidly create macroevolutionary novelty. Indeed, if the Eldredge and Gould and Stanley model of Punctuated Equilibrium were correct, they would need a model of evolution
that could create relatively rapid changes. Allometric
growth rates could cause the huge antlers of the Irish elk,
the single-toed horse, and the remarkable cerebral cortex of
Homo sapiens. By heterochrony, one could generate de
Beers patterns of neoteny and paedogenesis, which could

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

363

A New Evolutionary Synthesis

generate new and successful phenotypes. Goulds 1977 book

is important in exorcising the ghost of Haeckel and allowing
development to become part of evolutionary theory again.
Moreover, although heterochrony and allometry have
proved to be insufficient as mechanism to effect the integration of development and evolution (Raff and Kaufman, 1981;
Raff, 1996), it did focus attention on homology. By the
1980s, homology had become reestablished as a major area
in paleontology.
The molecular rediscovery of homology was predicated
on nucleic acid hybridization and protein sequencing. These
techniques showed that there were similarities between
protein sequences (such as those in globin) and that nucleic
acids also showed regions of similar or identical sequence.
Finessing the classic distinction between analogy and homology, Roy Britten (1967) proposed that homology between nucleic acid sequences referred to the degree of similarity between the nucleic acid sequences of different species. This was best observed when globin gene and protein
sequences were compared both within an organism and between organisms. Thus, the human a, b, d, gA, and gG
globins each shared certain sequences, but were different
in certain ways. Moreover, between species, the various
globins were also similar and the similarity was proportional to the relatedness of the species. Horse and human
a globins are distinct, differing in only 17 amino acids of
141. The only difference between human and gorilla a globin occurs at the 23rd amino acid. Since the various globins
(and their genes) are similar in structure within the body,
they can be said to be serially homologous (paralogous, to
use Fitchs 1970 term). Since they also are similar between
species, they would conform to Owens (1848) notion of
special homology (Fitchs orthologous category). The general homology of globins to one another implies knowledge
of the relationship of their structure to a particular function,
and the oxygen transport function is seen to be dependent
upon particular conserved sequence structures. Indeed, one
also finds this homologous structure in the proteins and
genes of myoglobin. Thus, as Jukes (1968) put it, the genes
responsible for the production of the globin portion of the
hemoglobins and myoglobins are all derived from a common archetypal piece of DNA, probably containing 486 base
pairs.
Not only does this language harken back to that of Owen,
but so does the mechanism for the production of the homologous sequences (Gilbert, 1980). Owen (1848) viewed the
archetypal vertebra as undergoing vegetative repetition
to produce a chain of identical vertebrae. Each of these vertebrae could then undergo independent modification for
its offices of existence. According to Britten and Kohnes
hypothesis (1968) for the generation of families of related
DNA sequences, there would be a nonrepeating archetypalsequence that would undergo saltatory replication
to form a tandem family of identical DNA sequences.
Thereafter, these duplicated copies would be free to undergo
independent mutation and be so selected. The molecular

biologists at the Carnegie Institute of Washington had rediscovered homology in the DNA.
But the most far reaching rediscovery of homology came
in developmental biology. The first series of discoveries
came from the study of homeosis. Homeotic traits had been
a sidelight of evolutionary theory ever since William Bateson collected them together in 1894. He noted that differences in the number or type of segment represented discontinuous patterns of evolution. In the 1940s, both C. H. Waddington and R. B. Goldschmidt identified mutations
whereby one type of insect segment was transformed into
another type of segment, and they claimed that these homeotic mutants might be the key for understanding the
relationship between genetics, development, and evolution.
In some of these mutants, parts of the antenna were replaced
by the homologous part of the leg; e.g., the tip of the antenna
was replaced by the claw of the leg. In other mutants, the
entire antenna had been replaced by the leg. In some mutants, the balancer (halteres) of the fly had been replaced by
wings, causing the di-pteran to resemble a more primitive
four-winged insect.
E. B. Lewis (1978, 1985) proposed a hypothesis that
brought these mutations to bear on evolution. It was based
on a notion of evolution by gene duplication (Ohno, 1970)
which, itself, had similarities to the homology theories of
Owen. According to Lewis, the second thoracic segment
(having both wings and legs) is the evolutionary baseline
for the insects. He then proposed that this gene should have
undergone several rounds of duplication and that there
should be one gene for each segment below the second thoracic level. As each successive gene became active, a new
set of structures are formed, distinguishing that segment
from any other. In the last segment, all these genes would
be active. Mutations in these genes could produce evolutionarily atavistic phenotypes, such as when those mutant
genes in the third thoracic segment convert the halteres
into wings.
Three major groups (E. B. Lewis and D. S. Hogness in
California; W. Gehring in Basel; T. Kaufman in Indiana; and
their respective students) used the new molecular techniques to isolate and sequence these genes, and they discovered a remarkably stable region: a 180-bp consensus sequence called the homeobox. It appeared that the genes
responsible for homeotic transformations were themselves
homologous. In the 1980s, another advance was made.
These same homoeotic genes were found to exist in vertebrates, initially in Xenopus laevis and then in mice, humans, birds, and fish. The original paper demonstrating vertebrate homeobox genes (Carrasco et al., 1984) noted that
if the frog gene cloned here eventually turns out to have
functions similar to that of the fruit fly genes, it would
represent the first development-controlling gene identified
in vertebrates. These genes were said to be homologous,
and since the homeotic genes appeared to create the anteriorposterior axis in flies, it was speculated that the same
genes might create the anteriorposterior axis in humans.
To some, this idea seemed bizarre. Vertebrate body segmen-

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

364

Gilbert, Opitz, and Raff

tation and insect segmentation are thought to be independently evolved modifications. Insects dont have somites or
bones. Vertebrates dont have germ bands or cuticles. It
seemed that the molecular biologists had forgotten the distinction between homology and analogy. Then, something
happened. First, it was shown that the homeotic genes of
mice, humans, and other vertebrates are arranged in the
same order on the chromosome as the homeotic gene complex in the fly. Second, it was shown that the anterior
posterior expression pattern of the individual genes was
the same in the fly and in vertebrates (see McGinnis and
Krumlauf, 1992; Krumlauf, 1993; Bachiller et al., 1994). And
last, it was shown that the enhancer region of a human
homeotic gene, such as deformed, can function within Drosophila to activate gene expression in the same relative
position as in the human embryoin the head (McGinnis
et al., 1990; Malicki et al., 1992).
In the 1990s, the use of homologous recombination to
functionally delete homeotic genes in mice enabled numerous laboratories to see what happened when vertebrates
lacked one or more of these genes. The results demonstrated
that these genes controlled the formation of the anterior
posterior axis in vertebrates as well as in flies and that
deletions of these genes could produce atavistic changes
such as the formation of reptilan jaw and neck vertebrae in
mice (Chisaka and Capecchi, 1991; Rijli et al., 1993). Studies by Gaunt (1994) and by Burke and her colleagues (1995)
have shown that the specific expression pattern of these
homeotic genes is responsible for forming the identities of
the vertebrae along the anteriorposterior axis in amniotes.
Indeed, the finding that every animal has similar genes, has
them in the same chromosomal order, and uses them to
specify the same relative positions along the anteriorposterior axis has caused Jonathan Slack and his colleagues
(1993) to go back even farther than Owen, to Etienne Geoffroy St-Hilaire, who felt that all animals were variations on
the same general plan of existence. At a particular phylotypic stage of development, each animal expresses these
genes to create the specification of its cells along the anteroposterior axis. This view stresses the similarities of embryonic development across the phyla. Even though insects
and vertebrates create their body axes, limbs, and nervous
systems in different ways, there appears to be an essential
underlying unity operating in the development of every animal on this planet. The comparison of the homeotic gene
complex to the Rosetta stone (Riddihough, 1992; Slack and
Tannahill, 1992) is apt: Their homologies enable us to translate our knowledge of Drosophila development into the unknown realm of vertebrate embryogenesis.
The segmentation of Drosophila and the segmentation of
vertebrates had been a classic example of analogy. Yet, here
it was seen as being directed by a homologous set of genes.
This demonstration of homologous genes for analogous
processes and structures has wreaked havoc with our definitions of analogy and homology. The insect eye and the
vertebrate eye are two examples of structures said to be
analogous. However, they can be shown to both be based

on the expression of the Pax-6 gene (Quiring et al., 1994),

and it is probable that the vertebrate and insect (and cephalopod) eyes are the modified descendents of a basic metazoan photoreceptive cell that was regulated by Pax-6. It has
recently been proposed (Chisholm and Horvitz, 1995) that
the Pax-6 family initially functioned to pattern part of the
head region (i.e., working as part of the anterior head field)
and only subsequently evolved more specific sensory functions. Similarly, the Xenopus gene chordin and the Drosophila gene short-gastrulation have similar sequences and
expression patterns, and they act similarly in vertebrate and
insect gastrulation (to counter the lateralizing effects of
BMP-4/decapentaplegic). Even though the types of gastrulation do not appear similar to any marked degree, the genes
controlling them may be homologous (Francois and Bier,
1995; Holley et al., 1995). Similarly, the heart of vertebrates
and the heart of insects have hardly anything in common
except their ability to pump fluids. Yet, they both appear
to to be predicated upon the expression of the same gene,
Csx/tinman (see Manak and Scott, 1994).
This gets us into an newly discovered and fascinating
realm of homologythe homology of process (Gilbert,
1996b). Whereas classic homology has been one of structurebe it of skeletons or genesthe homology of process
goes into the very mechanisms of development. Whereas
classical homology looks at the similarities between entities, the homology of process concerns the similarities of
dynamic interactions. The result is that although organs
(such as the vertebrate and arthropod eye, the vertebrate
and arthropod leg, etc.) can be structurally analogous, they
may be formed by processes that are homologous!
One of the best examples of such a process is the receptor
tyrosine kinase-ras signal transduction pathway that has
recently been found in mice, nematodes, and fruit flies. In
Drosophila, the determination of the photoreceptor seven
is accomplished when the sevenless protein (on the presumptive photoreceptor 7) binds to the bride of sevenless
protein (boss) on photoreceptor 8. This interaction activates
the tyrosine kinase of the sevenless protein to phosphorylate itself. The DRK protein then binds to these newly phosphorylated tyrosines through its src-homology-2 (SH2) region and activates the son of sevenless (SOS) protein. This
protein is a guanosine nucleotide exchanger and exchanges
GDP for GTP on the Ras1 G protein. This activates the G
protein, enabling it to transmit its signal to the nucleus
through the MAP kinase cascade. This same system has
been found to exist in the determination of the nematode
vulva, the mammalian epidermis, and the Drosophila terminal segments. The similarity in these systems is so striking that many of the components are interchangeable between species. The gene for human GRB2 can correct the
phenotypic defects of sem-5-deficient nematodes, and the
nematode sem-5 protein can bind to the phosphorylated
form of the human EGF receptor (see Greenwald and Rubin,
1992; Gilbert, 1994b). The process is thus historically (specifically) homologous between species (Drosophila retina/

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

365

A New Evolutionary Synthesis

nematode vulva) and serially homologous within species

(Drosophila retina/Drosophila acron and telson).
Another important pathway involves the Drosophila
wingless and hedgehog proteins. These proteins were found
to be critical in the formation of segmental boundaries in
the Drosophila embryo and of compartmental boundaries
in the larval imaginal discs. During the formation of the
parasegmental border of the embryo, the more posterior cell
secretes the hedgehog protein. This protein binds to a receptor on the anterior cell and stimulates the production of the
wingless protein. The wingless protein acts in a paracrine
fashion to inhibit the zest-white 3 kinase in the neighboring
cell. The inhibition of zw3 kinase releases the repression
of the hedgehog gene, thus stabilizing the pathway. The
winglesshedgehog system is serially homologous in Drosophila, being used later in the eye, leg, and wing imaginal
discs to specify the proximodistal axis (see Wilder and Perrimon, 1995). This system is also historically homologous.
In vertebrates, there are several homologues to wingless,
namely, the wnt proteins; the homologue to zest-white 3
is glycogen synthase kinase 3b (GSK-3b); and there are numerous hedgehog analogues, such as sonic hedgehog. In
vertebrates, the winglesshedgehog system is thought to be
needed for producing the body axes (as in Drosophila) and
the limbs (as in Drosophila). (Niswander et al., 1994, Ingham, 1994; Laufer et al., 1994.
So we not only have homologous genes, but homologous
pathways in organisms as diverse as flies, frogs, and yeasts.
We have come a long way from the time when Mayr (1966)
could state concerning macroevolution: Much that has
been learned about gene physiology makes it evident that
the search for homologous genes is quite futile except in
very close relatives.

The Rediscovery of the Morphogenetic Field

When we look at the homology of process, we notice
something else, as well. These interactions occur within
particular collections of cells that had formerly been identified as being fields. These domains, the limb field, the eye
field, the otic field, etc., were each isolatable, transplantable, and well-characterized landmarks on the embryo. In
some areas of developmental biology, the concept of the
field has persisted, and the notions of limb fields and heart
fields are still in the literature (see Sater and Jacobson, 1990;
Easton et al., 1994; Cohn et al., 1995). In such instances,
no claims are usually made other than that these areas of
mesoderm are destined to form these particular structures.
In recent years, several developmental biologists have revitalized the ideas of fields and have reclaimed their fundamental importance for both development and evolution.
Brian Goodwin (1982, 1995) has formulated a concept of a
morphogenetic field whose nongenic mechanisms of action
and wholism probably correspond quite well to the classical
notions of Paul Weiss and Alexander Gurwitsch. However,
this is a field that is outside developmental genetics and is
actively opposed to gene action as being important in field

functions. De Robertis and his associates (1991) synthesized

molecular and classical material to increase awareness
among modern developmental biologists of the old concepts
of morphogenetic gradient fields. At that time, however,
the interactions between parts of any field were still unknown, but De Robertis et al. (1991) emphasized the roles
that homeobox genes may play in initiating and organizing
these fields. Especially important to them were two observations concerning gradients produced by Hox proteins in
limb buds. The first was that gradients of these proteins
could induce the production of specific proteins at specific
sites and that these proteins may establish the conditions
for a field (such as the limb field or feather bud field) to
emerge. The second notion was that the gradients of these
proteins might establish the polar axes of these organs. Until recently, the interactions that constituted these fields
could not be identified. However, the discovery of the homologous pathways of development has given us new insights into how these fields are established and maintained.
Molecular biologists have recently rediscovered fields in
Drosophila. The imaginal discs of insects have long been
considered as gradient fields (see French et al., 1976; Ingham
and Martinez Arias, 1992; Williams et al., 1994), since they
are well-defined groups of cells whose interactions form an
organ, since they regulate to replace missing parts, and since
they retain their ability to generate the particular organ
when the disc is transplanted to other sites in the larva.
The work from Cohens and Carrolls laboratories is giving
us a fascinating picture of how interactions within these
fields create the leg is created and how changes in these
interactions can cause altered morphologies (Diaz-Benjumea et al., 1994; Williams et al., 1994). The Drosophila leg
field appears to be established by a rectilinear coordinate
system whereby the Hom/Hox genes (Scr, Antp, Ubx) determine the anteroposterior zone of competence to form legs,
whereas decapentaplegic expression is needed in the dorsoventral plane. The polarity of the leg is produced from the
interaction of three compartments within the disc. The posterior compartment is defined by the synthesis of the engrailed protein and the secretion of hedgehog protein. The
anterior dorsal compartment contains cells capable of producing decapentaplegic protein, and the anterior ventral
compartment contains cells competent to express wingless.
Upon induction by hedgehog protein, the band of dorsal
cells immediately anterior to the posterior border synthesize decapentaplegic protein, while the ventral cells immediately adjacent to the posterior border produce the wingless
protein. Those cells at the border of decapentaplegic and
wingless expression regions are instructed to produce the
distal-less protein, and these cells become the distalmost
portion, the claw, of the leg. In this way, the anteriorposterior and dorsalventral compartments of the disc create the
proximaldistal axis of the leg.
Few people would have expected that a similar situation
would exist for another embryological fieldthe vertebrate
limb field. After all, here is the classic example of analogy
as opposed to homology. The insect and vertebrate legs

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

366

Gilbert, Opitz, and Raff

share the same function, but thats about it. The insect leg
forms from the telescoping of the ectodermal imaginal disc.
The vertebrate limb forms from the reciprocal induction of
the Apical Ectodermal Ridge, the mesodermal Progress
Zone mesenchyme, and the mesodermal Zone of Polarizing
Activity. The insect leg has a cuticular exoskeleton, the
vertebrate limb has a complex endoskeleton of cartilage
and/or bone. There does not seem to be much in common.
The vertebrate limb field is thought to be initiated by the
localized secretion of fibroblast growth factor proteins
(Cohn et al., 1995). This, in turn, may be based on Hom/
Hox genes such as Hoxc-6 (which is present in the lateral
plate mesoderm at the place where the forelimb bud will
be formed; Oliver et al., 1988; De Robertis et al., 1991) or
Hoxb-8 (which is present in the posterior forelimb bud and
whose duplication in the anterior of that bud leads to mirror-image duplications of the posterior forelimb; Charite et
al., 1994). Recently, several laboratories have shown that
the same proteins that generate the insect leg also generate
the vertebrate limbs. Just as hedgehog protein from the posterior portion of the insect leg disc activates the decapentaplegic gene, so sonic hedgehog protein from the ZPA
mesoderm in the posterior of the bud activates BMP-2, a
vertebrate analogue of decapentaplegic (Francis et al., 1994).
In addition, the expression of sonic hedgehog is activated by
the diffusion of the wnt-7a protein (i.e., a wingless homolog)
from the dorsal ectoderm (Yang and Niswander, 1995; Parr
and McMahon, 1995). The molecular interactions within
the field needed to create a zone of polarizing activity, to
create an apical ectodermal ridge, and to create a progress
zone mesoderm are now becoming known, and they resemble the interactions that create the axes of the insect limb.
It seems that nature only figured how to make appendages
once. Moreover, nature seems to like to use the same pathways over and over again in different fields to make different
organs. The same decapentaplegic/hedgehog/wingless system appears to be working in the Drosophila eye-antennal
disc, where the conjunction of ventral wingless, dorsal dpp,
and posterior hh cause the synthesis of distal-less protein.
It is assumed that the targets of these proteins are different
in different discs, so that the genes for the appropriate organs are activated.
The concept of fields was also rediscovered by clinical
geneticists. Given that a specific malformation (such as an
extra thumb) can be caused by different mutations and be
a component of different syndromes, it was established that
the complex of anatomic structures that was malformed
together constituted a dysmorphogenetically reactive unit.
It was presumed that the same complex of anatomic structures constituted a morphogenetically reactive unit under
normal circumstances. The dysmorphogenetically reactive
fields defined on the basis of clinical syndromes were seen
to be the equivalent of the self-organizing, spatially coordinated, and temporally synchronized morphogenetic fields
of classical embryology. This equation was supplemented
by the observations that in many vertebrate species, the

same malformations (cyclopia, polydactyly, etc.) could be

produced experimentally or by mutation (Opitz, 1985).
The above-mentioned fields are the so-called secondary
fields. The primary field is the entire embryo during blastogenesis, before axis or cell determination. This is the field
that guarantees that each egg produces one (and only one)
embryo, despite the fact that many of the early blastomeres
can produce an entire embryo if isolated from the others
(Driesch, 1892; Hertwig, 1894; Spemann, 1919; Spratt and
Haas, 1960). This primary morphogenetic field was rediscovered by several groups of embryologists, including developmental biologists who demonstrated that mutual inhibitory interactions were occurring between the embryonic cells (Henry et al., 1989; Khaner and Eyal-Giladi, 1989;
Khaner and Wilt, 1991), clinical geneticists who postulated
such a field on the basis of clinical malformations (Opitz,
1993), and theorists (Raff et al., 1991) who predicted such
a global morphogenetic field on the evidence from development wherein evolutionary changes could occur only at
particular times in the life cycle.
The molecular analysis of the primary morphogenetic
field in Xenopus uncovered once again the activity of the
wnt genes (McMahon and Moon, 1989; Pierce and Kimelman, 1995; He et al., 1995). According to the wnt signaling
pathway, wnt acts to suppress activity or synthesis of the
zw3/GSK-3b gene product. If the pathway were blocked
such that GSK-3b is insensitive to inhibition by wnt, no
primary axis is formed. Even more interestingly, when
GSK3b is completely removed (by molecular means), two,
three, and even four dorsal axes form in the frog embryo.
One can also obtain frogs with multiple axes by adding
excess wnt mRNA. It appears, then, that the wnt pathway
is critical for maintaining embryonic individuality.

SUMMARY
We can now integrate these ideas together into the beginnings of a theory that includes homology, macroevolution,
and the developmental genetics of morphogenetic fields.
Morphogenetic fields assume the primary organizing activity here, as well as in the embryo.
1. Fields are discrete units of embryonic development.
They are produced by the interactions of genes and gene
products within specific bounded domains. They are therefore defined in terms of information that becomes translated
into spatial entities. Fields can be limited by diffusion, competence, gap junctions, or cell adhesion molecules. Changes
in these properties of the field result in changes in phenotype and lead to evolutionary novelty. Other changes within
the fields (such as those involving changes in the amount,
type, or duration of gene products or those involving mutations that alter the specificity DNA-protein binding) can
also cause evolutionary alterations.
2. Morphogenetic fields are modular entities. This modularity is an important key to biological order. The informa-

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

367

A New Evolutionary Synthesis

tion content or determinacy of a complex anatomical structure is orders of magnitude higher than that of the genome,
and such order rises from the use of standard parts, which
are arranged hierarchically, and which can interact with
each other (Riedl, 1977). Embryonic modules such as morphogenetic fields and organ rudiments are genetically specified, have autonomous attributes and hierarchical organization, and can change with regard to location, time, and interactions with other modules (Raff, 1996). Thus, a dynamic
modular structure is characteristic of metazoan organisms
and is a property of fields as well.
3. Although located in the same places, these rediscovered morphogenetic fields are not the same fields as those
postulated by Gurwitsch, Spemann, or Weiss. The older
morphogenetic fields were anatomically and cytoplasmically defined entities that were innocent of genes. The new
conceptions of morphogenetic fields are based on genetically defined interactions among cells, and the limits of
competence can be established by homeotic genes. The
high-in-the-herarchy genes, such as those encoding transcription factors Pax-6 and Lim1, most likely act to establish such fields.
4. Homologous morphogenetic fields can exist within
the same organism (serial process homology) or between
different organisms (orthologous process homology). An example of serial process homology include the ras pathway
in the retinal fields and the terminal segment fields of Drosophila. This pathway is orthogonally homologous to the
epidermal differentiation pathway in mammals. The expression of the Hom/Hox genes across the anteriorposterior axis of vertebrate and insect embryos would also constitute an orthologous homologous field between species, and
the use of the same genes in mice or chicks to generate the
dorsoventral axis of the limb would constitute a serially
homologous field in those organisms. Evolution depends on
the replication and modification of morphogenetic fields.
This may be seen in the origins of novel structures (insect
jaws, turtle carapace, butterfly wing eyespots) using the existing limb fields (Burke, 1989; Panganiban et al., 1994; Carroll et al., 1994). The mechanisms by which fields can be
replicated and then altered is a new area of research which
should produce new insights into the mechanisms of evolution (Jernvall, in press; Nijhout and Paulsen, personal communication).
5. Homologous genes/proteins can play different roles in
different fields. For example, sonic hedgehog activates different proteins in different fields. It works as a ZPA morphogen
within the limb field and as the inducer of floorplate and
motor neuron differentiation in the neuraxis field. Similarly,
the rel protein pathway is used in insects to establish the
ventral mesodermal cells (through the separation of the dorsal protein from the cactus protein), while vertebrates use
a homologous pathway of homologous genes for activating
immunoglobulin production (through the separation of NFkB from IkB) (Kidd, 1992; Shelton and Wasserman, 1993).
6. The fields explain pleiotropic and polytopic syndromes. In syndromes wherein a single mutant gene (or

gene pair) produces multiple congenital anomalies, several

organs may be affected together. In polytopy (Opitz, 1993),
organs are affected that are linked together in some intercellular developmental pathway. Here, for example, we see
that the renal-limb deficiencies might be explained because
early in development, they belong to a common field
(Dieker and Opitz, 1969). Paracrine factors from the mesonephros (probably insulin-like growth factor I) are needed
to promote the initial growth of the limb bud (Geduspan and
Solursh, 1992, 1993). Similarly, renal and gonadal tissues
are often affected together, since these organs constitute a
common field early in development. The syndromes such
as DiGeorge syndrome, CATCH-22 syndrome, and MEN2A, each have their disparate symptoms joined by a defect
in the neural crest fields of the mammalian embryo (see
Scambler, 1994). We also know that different mutations can
create the same phenotype by affecting the same field. Thus,
overexpression of wnt and deletions of zeste--white 3 give
the same or related phenotype. Multiple malformations occuring in syndromes can also be caused by true (mosaic)
pleiotropy (Hadorn, 1961; Gruneberg, 1962) wherein the organs cannot be joined together in a common field. In these
cases, the same molecule is thought to be used in several
different fields. Since msx-2 genes are expressed in developing limbs and teeth, we would expect a deficiency in
msx-2 to result in deformities of these two structures, even
though there is no connection between these two organs as
they develop. Indeed, the deficiency of msx-2 in humans
leads to a condition characterized by such abnormalities
(Jabs et al., 1993).
7. The field acts like an ecosystem (Weiss), and the deletions of certain genes can be regulated for under certain conditions. For example, the deletion of myoD in muscle cells
does not lead to marked deficiencies since within the field,
it represses a similar gene, myf-5. When MyoD is absent,
myf-5 is no longer repressed and can function like MyoD
(Rudnicki et al., 1993). This accounts for the buffering
noted by Waddington and the redundancy (belt and suspenders) noted by Spemann. The morphogenetic field thus unites
the atomism of the genetic and biochemical pathways within
the wholism of the developmental pathway.
8. The gene effects morphogenesis by operating within
the field; it has to work in concert with other genes in order
to function. It has long been known that the same gene
inherited through different generations can become expressed severely or benignly depending on its background. Freire-Maia (1975), for example, reports that
within one family, a mutant gene caused limb abnormalities ranging from severe phocomelia to a mild abnormality
of the thumb. This can be also be seen in malformations of
gonad formation. The Y-linked SRY gene needed for testis
morphogenesis of one particular strain of Mus musculus
cannot function to produce testes when placed into a different strain of the same species (Eicher and Washburn, 1983).
Similarly, the SRY gene of the AKR strain of mouse is not
able to produce a testis when placed into a C57 strain. Another gene, on chromosome 17 of the C57 strain, cannot

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

368

Gilbert, Opitz, and Raff

cooperate with the AKR Y chromosome (Eicher and Washburn, 1989). Thus, while each gene is perfectly wild-type
within its own strain, it acts a a deficient mutant when
placed in a different background.
9. Cooperation between inducer and responder cell types
is critical, and changes in these inducers or responders can
alter development. This can happen in several ways. First,
there can be a transfer of competence. If a pathway is
established such that a receptor binds a ligand and initiates
the cascade once the ligand is bound, then the pathway can
be activated by a different molecule if the receptor changes.
This can be accomplished experimentally, and it can explain phenomena such as Waddingtons genetic assimilation, I. I. Schmalhausens stabilizing selection, and
G. G. Simpsons Baldwin effect. There are several candidates for this occurring during evolution (see Gilbert, 1994b;
Sommer and Sternberg, 1994). Second, if the receptor or if
the inducer changes its level or duration of activity, it can
alter the morphology of the organ. For instance, if the receptors for a growth factor stayed active for one more cell division, or if the cells secreting the growth factor produced
these factors for longer periods of time while the responding
cells remained competent, then the organ would be greatly
enlarged. It is possible that heterochronies and allometries
can be produced in this fashion. Third, the requirement for
cells to interact within a field could be a mechanism for
speciation. The evolution of receptorligand systems such
as those studied for bindin (Hofmann and Glabe, 1994) and
growth hormones (Moyle et al., 1994) may be crucial for
discussions of how species diverge from common ancestors.
10. Just as the cell is seen to be the unit of structure and
function in the bodynot the genes that act through it
so the morphogenetic field can be seen as a major unit
of ontogenetic and phylogenetic change. In declaring the
morphogenetic field to be a major module of developmental
and evolutionary change, we are, of course, setting it up as
an alternative to the solely genetic model of evolution and
development. This, however, is not to be seen as antagonistic to the principle that genes are important in evolution
or development. This is not in any way denied. But just as
the genes make the cells and the cells form the body, so the
gene products first need to interact to create morphogenetic
fields in order to have their effects. Changes in these fields
then change the ways that animals develop.

CODA
In 1895, Wilhelm Roux, in his manifesto for experimental
embryology, postulated that there would be two types of
developmental mechanics. The firstontogenetic developmental mechanicswould uncover how development occurred. The secondphylogenetic developmental mechanicswould determine how changes in embryonic development caused evolutionary change. A century later, we are
starting to make good on Rouxs prophecy. The homologies

of process within morphogenetic fields provide some of the

best evidence for evolutionjust as skeletal and organ homologies did earlier. Thus, the evidence for evolution is
better than ever. The role of natural selection in evolution,
however, is seen to play less an important role. It is merely
a filter for unsuccessful morphologies generated by development. Population genetics is destined to change if it is not
to become as irrelevant to evolution as Newtonian mechanics is to contemporary physics. The population genetics of
regulatory genes and their possible combinations within
fields should become a major new research program. Developmental genetics would also change, reflecting an emphasis on the initiation and maintainance of genetic circuits
within cells and epigenetic circuits within the field. One of
its major research programs would be to find the target
genes of these pathways which differ from field to field and
from organism to organism, i.e., those genes that provide
the diversity in evolution.
Developmental biology is reclaiming its appropriate place
in evolutionary theory. We conclude with a remarkable
prophecy from one of those evolutionary-minded embryologists, Gavin de Beer (1951), who saw homology and fields
as being crucial to the study of evolution:
But since phylogeny is but the result of modified ontogeny, there
is the possibility of a causal analytic study of present evolution in
an experimental study of the variability and genetics of ontogenetic
processes. Finally, it may be possible that, freed from the trammels
and fetters which have so long confined thought, the whole of the
animal kingdom may appear in a new light, more homogeneous
and compact than had been imagined, and with the gaps between
its major groups less formidable and perhaps even bridgeable.

ACKNOWLEDGMENTS
The authors acknowledge the constructive criticisms of Drs. I.
Dawid, E. M. De Robertis, and J. Jernvall on earlier drafts of the
manuscript.

REFERENCES
Adams, M. B. (1990). La genetique des populations: Etait-elle une
genetique evolutive? In Histoire de la Genetique: Pratiques,
Techniques et Theories (J.-L. Fischer and W. H. Schneider, Eds.),
pp. 152171. Editions Sciences, Paris.
Alberch, P. (1982). Developmental constraints in evolutionary processes. In Evolution and Development: Report of the Dahlem
Workshop on Evolution and Development (J. T. Bonner, Ed.),
pp. 313332. Springer-Verlag, New York.
Allen, G. E. (1978). Life Science in the Twentieth Century. Cambridge Univ. Press, NY.
Ayala, F. J. (1983). Microevolution and macroevolution. In Evolution from Molecules to Men (D. S. Bendall, Ed.), pp. 387402.
Cambridge Univ. Press, Cambridge.
Bachiller, D., Macias, D., Duboule, D., and Morata, G. (1994). Con-

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

369

A New Evolutionary Synthesis

servation of a functional hierarchy between mammalian and insect Hox/HOM genes. EMBO J. 13, 19301941.
Baldwin, J. M. (1902). Development and Evolution. Macmillan
Co., New York.
Bateson, W. (1894). Materials for the Study of Variation, Treated
with Especial Regard to Discontinuity in the Origin of Species.
Macmillan Co., New York.
Beadle, G., and Ephrussi, B. (1937). Development of eye colors in
Drosophila: Diffusible substances and their interrelations. Genetics 22, 7686.
Beatty, J. (1994). Opportunities for genetics in the atomic age. Paper
presented at the Mellon Symposium, Institutional and Disciplinary Contexts of the Life Sciences, Massachusetts Inst. Technol., Cambridge, MA.
de Beer, G. R. (1940). Embryos and Ancestors. Oxford Univ. Press,
Oxford.
de Beer, G. R. (1951). Embryos and Ancestors, revised ed. Oxford
Univ. Press, Oxford.
Berrill, N. J. (1941). Spatial and temporal growth patterns in colonial
organisms. Growth (Suppl.) 5, 89111.
Boveri, T. (1910). Die Potenzen der Ascaris-Blastomeren bei abgeanderter Furchung, zugleich ein Beitrag zur Frage qualitativ-ungleicher Chromosomen-Teilung. In Festschrift fur Richard Hertwig, Vol. 3. Fischer, Jena.
Britten, R. J. (1967). Carnegie Inst. Wash. Yearbook 66, 68.
Britten, R. J., and Kohne, D. (1968). Repeated sequences in DNA.
Science 161, 529540.
Burke, A. C. (1989). Development of the turtle carapace: Implications for the evolution of a novel bauplan. J. Morphol. 199, 363
378.
Burke, A. C., Nelson, A. C., Morgan, B. A., and Tabin, C. (1995).
Hox genes and the evolution of vertebrate axial morphology. Development 121, 333346.
Carraco, A. E., McGinnis, W., Gehring, W. J., and De Robertis,
E. M. (1984). Cloning of an X. laevis gene expressed during embryogenesis coding for a peptide region homologous to Drosophila homeotic genes. Cell 37, 409414.
Carroll, S. B., Gates, J., Keys, D. N., Paddock, S. W., Panganiban,
G. E. F., Selegue, J. E., and Williams, J. A. (1994). Pattern formation and eyespot determination in butterfly wings. Science 265,
109114.
Charite, J., DeGraaff, W., Shen, S. B., and Deschamps, J. (1994).
Ectopic expression of Hoxb-8 causes duplications of the ZPA in
the forelimb and homeotic transformation of axial structures.
Cell 78, 589601.
Child, C. M. (1915). Individuality in Organisms. Univ. of Chicago
Press, Chicago.
Child, C. M. (1941). Patterns and Problems of Development.
Univ. of Chicago Press, Chicago.
Chisaka, O., and Cappecchi, M. R. (1991). Regionally restricted
developmental defects resulting from targeted disruption of the
homeobox gene hox 1.5. Nature 350, 473479.
Chisholm, A. D., and Horvitz, H. R. (1995). Patterning of the Caenorhabditis elegans head region by the Pax-6 family member vab3. Nature 377, 5255.
Cohn, M. J., Izpisua-Belmonte, J. C., Abud, H., Heath, J. K., and
Tickle, C. (1995). Fibroblast growth factors induce additional
limb development from the flank of chick embryos. Cell 80, 739
746.
De Robertis, E. A., Morita, E. M., and Cho, K. W. Y. (1991). Gradient
fields and homeobox genes. Development 112, 669678.
Diaz-Benjumea, F. J., Cohen, B., and Cohen, S. M. (1994). Cell inter-

action between compartments establishes the proximal-distal

axis of Drosophila legs. Nature 372, 175179.
Dieker, H., and Opitz, J. M. (1969). Associated acral and renal malformations. Birth Defects Orig. Artic. Ser. 5(3), 6877.
Dobzhansky, T. (1937). Genetics and the Origin of Species. Columbia Univ. Press, New York.
Dobzhansky, T. (1951). Genetics and the Origin of Species. 3rd
ed. Columbia Univ. Press, New York.
Driesch, H. (1892). The potency of the first two cleavage cells in
echinoderm development. Experimental production of partial
and double formations. In Foundations of Experimental Embryology (B. H. Willier and J. M. Oppenheimer, Eds.), pp. 3850.
Hafner, New York.
Easton, H. S., Armstrong, J. B., and Smith, S. C. (1994). Heart specification in the Mexican axolotl. Dev. Dyn. 200, 313320.
Edelman, G. (1988). Topobiology. Basic Books, New York.
Eicher, E. M., and Washburn, L. (1983). Inherited sex reversal in
mice: Identification of a new primary sex-determining gene. J.
Exp. Zool. 228, 297304.
Eicher, E. M., and Washburn, L. (1989). Normal testis determination in the mouse depends on genetic interaction of a locus on
chromosome 17 and the Y chromosome. Genetics 123, 173179.
Eldredge, N., and Gould, S. J. (1972). Punctuated equilibria: An
alternative to phyletic gradualism. In Models of Paleobiology
(T. J. M. Schopf, Ed.), pp. 82115.
Filipchenko, I. A. (1929). Izmenchivost i Metody ee Izucheniia,
4th ed. Gosidzat, Leningrad.
Fitch, W. M. (1970). Distinguishing homologous from analogous
proteins. Syst. Zool. 19, 99113.
Francis, P. H., Richardson, M. K., Brickell, P. M., and Tickle, C.
(1994). Bone morphogenesis proteins and a signal transduction
pathway that controlls patterning in the developing chick limb.
Development 120, 209218.
Francois, V., and Bier, E. (1995). Xenopus chordin and Drosophila
short gastrulation genes encode homologous proteins functioning in dorsal-ventral axis formation. Cell 80, 1920.
French, V., Bryant, P. J., and Bryant, S. V. (1976). Pattern regulation
in epimorphic fields. Science 193, 969981.
Freire-Maia, N. (1975). A heterozygote expression of a recessive
gene. Hum. Hered. 25, 302304.
Gasman, D. (1971). The Scientific Origins of National Socialism.
MacDonald, London.
Gaunt, S. J. (1994). Conservation in the Hox code during morphological evolution. Int. J. Dev. Biol. 38, 549552.
Geduspan, J. S., and Solursh, M. (1992). A growth-promoting influence from the mesonephros during limb outgrowth. Dev. Biol.
151, 242250.
Geduspan, J. S., and Solursh, M. (1993). Effects of the mesonephros
and insulin-like growth factor I on chondrogenesis of limb explants. Dev. Biol. 156, 177187.
Geoffroy-Saint-Hilaire, E. (1830). Principes de Philosophie Zoologique, discutes en Mars, 1830, au Sein de lAcademie Royale des
Sciences. Pichon et Didier, Paris.
Gilbert, S. F. (1978). The embryological origins of the gene theory.
J. Hist. Biol. 11, 307351.
Gilbert, S. F. (1980). Owens vertebral archetype and evolutionary
genetics: A Platonic appreciation. Perspect. Biol. Med. 23, 475
488.
Gilbert, S. F. (1988). Cellular politics: Just, Goldschmidt, and the
attempts to reconcile embryology and genetics. In The American Development of Biology (R. Rainger, K. Benson, J. Maiensch-

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

370

Gilbert, Opitz, and Raff

ein, Eds.), pp. 311346. Univ. of Pennsylvania Press, Philadelphia.

Gilbert, S. F. (1994a). Dobzhansky, Waddington, and Schmalhausen: Embryology and the Modern Synthesis. In The Evolution of Theodosius Dobzhansky (M. Adams, Ed.), pp. 143154.
Princeton Univ. Press, Princeton.
Gilbert, S. F. (1994b). Developmental Biology. 4th ed. Sinauer,
Sunderland, MA.
Gilbert, S. F. (1996a). Adaptive enzymes and the entrance of molecular biology into embryology. In The History of Molecular Biology (S. Sarkar, Ed.), Kluwer Academic, Dordrecht, in press.
Gilbert, S. F. (1996b). Cellular dialogues during development. In
Gene Regulation and Fetal Development: Procedings of the
Third International Congress on Fetal Genetic Pathology (G.
Neri, E. Martini-Neri, and J. M. Opitz, Eds.). WileyLiss, New
York, in press.
Gilbert, S. F., and M. Faber. (1996). Looking at embryos: The visual
and conceptual aesthetics of emerging form. In Aesthetics and
Science: The Elusive Synthesis (A. I. Tauber, Ed.). Kluwer Academic, Dordecht, in press.
Gluecksohn-Schoenheimer, S. (1938). The development of two
tailless mutants in the house mouse. Genetics 23, 573584.
Goldschmidt, R. B. (1940). The Material Basis of Evolution. Yale
Univ. Press, New Haven.
Goodwin, B. (1982). Development and evolution. J. Theoret. Biol.
97, 4355.
Goodwin, B. (1995). How the Leopard Changed its Spots: The
Evolution of Complexity. Scribners, New York.
Gottlieb, G. (1992). Individual Development and Evolution: The
Genesis of Novel Behavior. Oxford Univ. Press, New York.
Gould, S. J. (1977). Ontology and Phylogeny. Harvard Univ. Press,
Cambridge.
Gould, S. J. (1980). Is a new general theory of evolution emerging?
Paleobiology 6, 119130.
Gould, S. J. (1982). The meaning of punctuated equilibrium and its
role in validating a hierarchical approach to macroevolution. In
Perspectives in Evolution (R. Milkman, Ed.), pp. 83104. Sinauer, Sunderland, MA.
Gould, S. J., and Lewontin, R. C. (1979). The spandrels of San Marco
and the Panglossian paradigm: A critique of the adaptationist
program. Proc. R. Soc. London Ser. B 205, 581598.
Gregory, W. K. (1917). Genetics versus paleontology. Am. Nat. 51,
622635.
Greenwald, I., and Rubin, G. M. (1992). Making a difference: The
role of cellcell interactions in establishing separate identities
for equivalent cells. Cell 68, 271281.
Gruneberg, H. (1962). The Pathology of Development. Blackwell,
Oxford.
ber Determination, Normierung und
Gurwitsch, V. A. (1910). U
Zufall in der Ontogenese. Arch. Entwicklungsmech. Org. 30,
133193.
Gurwitsch, V. A. (1912). Die Vererbung als Verwirklichungsvorgang. Biol. Centralblatt. 32, 458486.
ber den Begriff des embryonalen Feldes.
Gurwitsch, V. A. (1922). U
Rouxs Arch. Entwicklungsmech. Org. 51, 383415.
Hadorn, E. (1961). Developmental Genetics and Lethal Factors
(U. Mittwoch, trans.). Wiley, New York.
Hall, B. K. (Ed.) (1994). Homology: The Hierarchical Basis of Comparative Anatomy. Academic Press, New York.
Hamburger, V. (1980). Embryology and the Modern Synthesis in
evolutionary theory. In The Evolutionary Synthesis: Perspec-

tives on the Unification of Biology (E. Mayr and W. Provine,

Eds.), pp. 97112. Cambridge Univ. Press, NY.
Haraway, D. J. (1976). Crystals, Fabrics, and Fields: Metaphors of
Organicism in Twentieth-Century Developmental Biology.
Yale Univ. Press, New Haven.
Harrison, R. G. (1918). Experiments on the development of the
forelimb of Amblystoma, a self-differentiating equipotential system. J. Exp. Zool. 25, 413461.
Harrison, R. G. (1937). Embryology and its relations. Science 85,
369374.
Harwood, A. J., Plyte, S. E., Woodgett, J., Strutt, H., and Kay, R. R.
(1995). Glycogen synthase kinase 3 regulates cell fate in Dictyostelium. Cell 80, 139148.
He, X., Saint-Jeannet, J.-P., Woodgett, J. R., Varmus, H. E., and
Dawid, I. B. (1995). Glycogen synthase kinase 3 and dorsoventral
patterning in Xenopus embryos. Nature 374, 617622.
Henry, J., Amemiya, S., Wray, G. A., and Raff, R. (1989). Early
inductive interactions are involved in restricting cell fates of the
mesomeres in sea urchin embryos. Dev. Biol. 136, 140153.
Herrmann, H. (1964). Biological field phenomena: Facts and concepts. In Form and Strategy in Science (J. R. Gregg and
F. T. C. Harris, Eds.), pp. 343362. Reidel, Dordrecht.
Hertwig, O. (1894). The Biological Probem of Today: Preformation
or Epigenesis? (P. C. Mitchell, trans.). Macmillan Co., New
York.
Hofmann, A., and Glabe, C. (1994). Bindin, a multifunctional sperm
ligand and the evolution of new species. Semin. Dev. Biol. 5,
233242.
Holley, S. A., Jackson, P. D., Sasai, Y., Lu, B., De Robertis, E. M.,
Hoffmann, F. M., and Ferguson, E. L. (1995). A conserved system
for dorsal-ventral patterning in insects and vertebrates involving
short gastrulation and chordin. Nature 376, 249253.
Huxley, J. (1932). Problems of Relative Growth. Dial Press, New
York.
Huxley, J., and de Beer, G. R. (1934). The Elements of Experimental Embryology. Cambridge Univ. Press, Cambridge.
Huxley, T. H. (1858). On the theory of the vertebrate skull. Sci.
Memoirs 1, 538606.
Ingham, P. W. (1994). Hedgehog points the way. Curr. Biol. 4, 1
4.
Ingham, P. W., and Martinez Arias, A. (1992). Boundaries and fields
in early embryos. Cell 68, 221235.
Jabs, E. W., and 11 others. (1993). A mutation in the homeodomain
of the human MSX2 gene in a family affected with autosomal
dominant craniosynostosis. Cell 75, 443450.
Jernvall, J. (1995). Mammalian molar cusp patterns: developmental
mechanisms of diversity. Thesis. Acta Zool. Fennica 198, 161.
Jukes, T. (1968). Molecules and Evolution. Columbia Univ. Press,
New York.
Kettlewell, H. B. D. (1959). Darwins missing evidence. Sci. Am.
200(3), 4853.
Khaner, O., and Eyal-Giladi, H. (1989). The chicks marginal zone
and primitive streak formation. I. Coordinative effect of induction and inhibition. Dev. Biol. 134, 206214.
Khaner, O., and Wilt, F. (1991). Interactions of different vegetal cells
with mesomeres during early stages of sea urchin development.
Development 112, 881890.
Kidd, S. (1992). Characterization of the Drosophila cactus locus
and analysis of interactions between cactus and dorsal proteins.
Cell 71, 623635.
King, M.-C., and Wilson, A. C. (1975). Evolution at two levels in
humans and chimpanzees. Science 188, 107116.

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

371

A New Evolutionary Synthesis

Krumlauf, R. (1993). Hox genes and pattern formation in the

branchial region of the vertebrate head. Trends Genet. 9, 106
112.
Lammer, E. J., and Opitz, J. M. (1986). The DiGeorge anomaly as
a developmental field effect. Am. J. Med. Genet. 2, 143150.
Laufer, E., Nelson, C. E., Johnson, R., Morgan, B. A., and Tabin, C.
(1994). Sonic hedgehog and Fgf-4 act through a signalling cascade
and feedback loop to integrate growth and patterning of the developing limb bud. Cell 79, 9931003.
Lewis, E. B. (1978). A gene complex controlling segmentation in
Drosophila. Nature 276, 565570.
Lewis, E. B. (1985). Regulation of the genes of the bithorax complex
in Drosophila. Cold Spring Harbor Symp. Quant. Biol. 50, 155
164.
Lillie, F.R. (1898). Adaptation in cleavage. In Biological Lectures
from the Marine Biological Laboratories, Woods Hole, Massachusetts. pp. 4367. Ginn, Boston.
Maienschein, J. (1991). Transforming Traditions in American Biology, 18801915. Johns Hopkins Univ. Press, Baltimore.
Malicki, J., Cianetti, L. C., Peschle, C., and McGinnis, W. (1992).
Human HOX4B regulatory element provides head-specific expression in Drosophila embryos. Nature 358, 345347.
Manak, J. R., and Scott, M. P. (1994). A class act: Conservation of
homeodomain protein functions. Development Suppl. 6171.
Mayr, E. (1966). Animal Species and Evolution. Harvard Univ.
Press, Cambridge.
McGinnis, N., Kuziora, M. A., and McGinnis, W. (1990). Human
Hox 4.2 and Drosophila Deformed encode similar regulatory
specificities in Drosophila embryos and larvae. Cell 63, 969976.
McGinnis, W., and Krumlauf,R. (1992). Homeobox genes and axial
patterning. Cell 68, 283302.
McMahon, A. P., and Moon, R. T. (1989). Ectopic expression of the
proto-oncogene int-1 in Xenopus embryos leads to duplication
of the embryonic axis. Cell 58, 10751084.
Mitman, G., and Fausto-Sterling, A. (1992). Whatever happened to
Planaria? C. M. Child and the physiology of inheritance. In The
Right Tool for the Right Job: At Work in Twentieth-Century Life
Sciences, (A. E. Clarke and J. H. Fujimura, Eds.), pp. 172197.
Princeton Univ. Press, Princeton.
Monod, J. (1971). Chance and Necessity. Knopf, New York.
Morgan, T. H. (1926a). The Theory of the Gene. Yale Univ. Press,
New York.
Morgan, T. H. (1926b). The genetics and physiology of development. Am. Nat. 60, 489515.
Morgan, T. H. (1932a). The rise of genetics. Science 76, 261288.
Morgan, T. H. (1932b). The Scientific Basis of Evolution. Norton,
New York.
Morgan, T. H. (1934). Embryology and Genetics. Columbia Univ.
Press. New York.
Moyle, W. R., Campbell, R. K., Myers, R. V., Bernard, M. P., Han,
Y., and Wang, X. (1994). Co-evolution of ligand-receptor pairs.
Nature 368, 251255.
Needham, J. (1931). Chemical Embryology. Cambridge Univ.
Press, Cambridge.
Needham, J. (1950). Biochemistry and Morphogenesis. Cambridge Univ. Press, Cambridge.
Niswander, L., Jeffrey, S., Martin, G. R., and Tickle, C. (1992). A
positive feedback loop coordinates growth and patterning in the
vertebrate limb. Nature 371, 609612.
Oliver, G., Wright, C. V. E., Hardwicke, J., and De Robertis, E. M.
(1988). A gradient of homeodomain protein in developing forelimbs of Xenopus and mouse embryos. Cell 55, 10171024.

Ohno, S. (1960). Evolution by Gene Duplication. Springer-Verlag,

New York.
Opitz, J. M. (1985). The developmental field concept. Am. J. Med.
Genet. 21, 111.
Opitz, J. M. (1993). Blastogenesis and the primary field in human
development. Birth Defects Orig. Artic. Ser. 29, 134.
Oppenheimer, J. M. (1966). The growth and development of developmental biology. In Major Problems in Developmental Biology (M. Locke, Ed.), pp. 127. Academic Press, New York.
Osborn, H. F. (1925). The origin of species as revealed by vertebrate
paleontology. Nature 115, 961963.
Owen, R. (1848). On the Archetype and Homologies of the Vertebrate Skeleton. Taylor, London.
Panganiban, G. E. F., Nagy, L., and Carroll, S. B. (1994). The role
of the distal-less gene in the development and evolution of insect
limbs. Curr. Biol. 4, 671675.
Parr, B. A., and McMahon, A. P. (1995). Dorsalizing signal Wnt-7a
required for normal polarity of D-V and A-P axes of mouse limb.
Nature 374, 350353.
Paul, D. B. (1988). The selection of survival of the fittest. J. Hist.
Biol. 21, 411424.
Pierce, S. B., and Kimelman, D. (1995). Regulation of Spemann
organizer formation by the intracellular kinase Xgsk-3. Development 121, 755765.
Quiring, R., Walldorf, U., Kloter, U., and Gehring, W. J. (1994).
Homology of the eyeless gene of Drosophila to the Small eye
gene in mice and Aniridia in humans. Science 265, 785789.
Raff, R. A. (1996). The Shape of Life. Genes, Development, and
the Evolution of Animal Form. Univ. of Chicago Press, Chicago,
in press.
Raff, R. A., and Kaufman, T. C. (1983). Embryos, Genes, and Evolution. Macmillan Co., New York.
Raff, R. A., Wray, G. A., and Henry. J. J. (1991). Implications of
radical evolutionary changes in early development for concepts
of developmental constraint. In New Perspectives in Evolution
(L. Warren and M. Meselson, Eds.), pp. 189207. WileyLiss,
New York.
Reidl, R. (1978). Order in Living Organisms: A Systems Analysis
of Evolution (R. P. S. Jefferies, trans.). Wiley, New York.
Riddihough, G. (1992). Homing in on the homeobox. Nature 357,
643644.
Rijli, F. M., Mark, M., Lakkaraju, S., Dierlich, A., Dolle, P., and
Chambon, P. (1993). A homeotic transformation is generated in
the rostral branchial region of the head by the disruption of Hoxa2, which acts as a selector gene. Cell 75, 13331349.
Roth, V. L. (1984). On homology. Biol. J. Linn. Soc. 22, 1329.
Roux, W. (1895). Einleitung. Rouxs Arch. Entwicklungsmech. Org.
1, 142. Translated (W. M. Wheeler) as The problems, methods,
and scope of developmental mechanics. Reprinted in Maienschein, J. (1986) Defining Biology, pp. 107148. Harvard Univ.
Press, Cambridge.
Rudnicki, M. A., Schnegelsberg, P. N. J., Stead, R. H., Braun, T.,
Arnold, H.-H., and Jaenisch, R. (1993). MyoD or myf-5 is required
in a functionally redundant manner for the formation of skeletal
muscle. Cell 75, 13511359.
Sander, K. (1994). Of gradients and genes: Developmental concepts
of Theodor Boveri and his students. Rouxs Arch. Dev. Biol. 203,
295297.
Sater, A. K., and Jacobson, A. G. (1994). Restriction of the heart
morphogenetic field in Xenopus laevis. Dev. Biol. 140, 328336.
Scambler, P. J. (1994). DiGeorge Syndrome and related birth defects.
Semin. Dev. Biol. 5, 303310.

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio

372

Gilbert, Opitz, and Raff

Schultz, J. (1935). Aspects of the relation between genes and development in Drosophila. Am. Nat. 69, 3054.
Shelton, C. A., and Wasserman, S. A. (1993). pelle encodes a protein
kinase required to establish dorsal-ventral polarity in the Drosophila embryo. Cell 72, 515525.
Slack, J. M. W., and Tannahill, D. (1992). Mechanism of anterioposterior specification in vertebrates: Lessons from amphibians.
Development 114, 285302.
Slack, J. M. W., Holland, P. W. H., and Graham, C. F. (1993). The
zootype and the phylotypic stage. Nature 361, 490492.
Smocovitis, V. B. (1994). Organizing evolution: founding the society
for the study of evolution (19391950). J. Hist. Biol. 27, 241
310.
Sommer, R. J., and Sternberg, P. W. (1994). Changes of induction
and competence during the evolution of vulva development in
nematodes. Science 265, 114118.
Spemann, H. (1919). Experimentelle Forschungen zum Determinations- und Individualitats-problem. Naturwiss. 7, 581591.
Spemann, H. (1921). Die Erzeugung tierischer Chimaeren durch
heteroplastiche embryonale Transplantation zwischen Triton
cristatus u. taeniatus. Rouxs Arch. Entwicklungsmech. Org. 48,
533570.
Spemann, H. (1938). Embryonic Development and Induction.
Yale Univ. Press, New Haven.
Spiegelman, S. (1948). Differentiation as the controlled production
of unique enzymatic patterns. In Growth in Relation to Differentiation and Morphogenesis (J. F. Danielli, and R. Brown, Eds.),
pp. 286325. Cambridge Univ. Press, Cambridge.
Spratt, N. T., Jr., and Haas, H. (1960). Integrative mechanisms in
the development of early chick blastoderm. I. Regulated potentiality of separate parts. J. Exp. Zool. 145, 97138.
Stanley, S. M. (1979). Macroevolution: Pattern and Process. Freeman, San Francisco.

Swett, F. A. (1923). The prospective significance of the cells contained in the four quadrants of the primitive limb disc of Amblystoma. J. Exp. Zool. 37, 207217.
Van Valen, L. (1982). Homology and causes. J. Morphol. 173, 305
312.
Waddington, C. H. (1940). Organisers and Genes. Cambridge
Univ. Press, Cambridge.
Waddington, C. H. (1966). Fields and gradients. In Major Problems
in Developmental Biology (M. Locke, Ed.), pp. 105124. Academic Press, New York.
Wagner, G. P. (1989). The origin of morphological characters and
the biological basis of homology. Evolution 43, 11571171.
Wake, D. (1994). Comparative terminology. Science 265, 268269
Weiss, P. (1923). Naturwissen. 11, 669 [quoted in Weiss, P. (1939).
Principles of Development Holt, New York].
Weiss, P. (1939). Principles of Development. Holt, New York.
Wilder, E. L., and Perrimon, N. (1995). Dual functions of wingless
in the Drosophila leg imaginal disc. Development 121, 477488.
Williams, J. A., Paddock, S. W., Vorwerk, K., and Carroll, S. B.
(1994). Organization of wing formation and induction of a wingpatterning gene at the dorsalventral compartment boundary.
Nature 368, 299305.
Wilson, E. B. (1896). The Cell in Development and Inheritance.
Macmillan Co., New York.
Wilson, E. B. (1898). Cell lineages and ancestral reminiscences.
In Biological Lectures from the Marine Biological Laboratories,
Woods Hole, Masssachusetts, pp. 2142. Ginn, Boston.
Yang, Y., and Niswander, L. (1995). Interaction between the signalling molecules WNT7a and SHH during vertebrate limb development: Dorsal signals regulate anteroposterior patterning. Cell 80,
939947.
Received for publication April 18, 1995
Accepted October 24, 1995

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

/ m4881f8085

01-09-96 20:04:25

dba

Dev Bio