Indeed, Cohn and Tickle (1999, p. 475) found Hoxc-8 and Hoxb-5 expression “throughout the python lateral plate mesoderm, with expression terminating at the very anterior limit of the trunk. Thus, the entire vertebral column anterior to the cloaca exhibits patterns of Hox gene expression consistent with thoracic identity, and we were unable to detect restricted Hox expression patterns in the lateral plate mesoderm associated with forelimb position in other tetrapods.” In interesting and confirming contrast, they detected a sharp posterior boundary of Hoxc-8 expression right at the level of the hindlimb rudiments, “which coincides with the last thoracic vertebra in older animals.” Their phyletic hypothesis underscores the power of constraining rules as positive channels that can be tweaked in rare and interesting ways to yield remarkable phenotypic and functional excursions into novelty — but always under the rubric of Hox rules and their potentiating, but also directional, flexibilities: “Expansion of these Hox gene expression domains in both paraxial and lateral plate mesoderm may be the mechanism which transformed the entire snake trunk towards a thoracic/flank identity and led directly to the absence of forelimb development during snake evolution” (1999, p. 475).
AN EPILOG ON DOBZHANSKY'S LANDSCAPE AND THE DOMINANT ROLE OF HISTORICAL CONSTRAINT IN THE CLUMPED POPULATION OF MORPHOSPACE. As I emphasized in setting out the varied categories and evolutionary significances of constraint (see pp. 1151–1161 and Figs. 10-10 and 10-11), the historical vertex treated in this Chapter 10 does not refute the functional or adaptational premises of traditional Darwinism by asserting a nonadaptive origin for the constraints thus generated — for this more direct challenge arises from the structural vertex that will be treated in Chapter 11. That is, I do not doubt that most, or nearly all, constraints from the historical vertex originate as direct adaptations in the ancestral taxon of their initial appearance. But, having thus originated, these adaptations may then “congeal” to limit directions of potential alteration in descendant taxa (the negative meanings), or to channel future change in preferred directions that often accelerate or grant easier access to adaptive solutions (the positive meanings). In terms of the classical model of Galton's polyhedron, the pool cue of natural [Page 1174] selection may always do the actual pushing, but if internal channels — set by history, and grafted into the genetic and developmental architecture of current organisms — designate a limited set of possible pathways as conduits for selection's pushing, then these internal constraints can surely claim equal weight with natural selection in any full account of the causes of any particular evolutionary change.
But if the challenge posed by historical constraint to traditional Darwinian functionalism does not lie in an argument about nonadaptive origins, then how can this category of constraint rectify and expand evolutionary theory beyond the narrowness imposed by overly adaptationist versions of Darwinism favored during the heyday of the Modern Synthesis (see Chapter 7)?
In describing my basic framework of argument, I asserted (see pp. 1055–1057) that the challenge of historical constraint resides in a “metaquestion” about the role of adaptation in establishing the dumpiness of occupied morphospace, not in a direct inquiry about the adaptive status of each evolutionary novelty considered one-by-one. In short, I argued that the markedly inhomogeneous occupation of morphospace — surely one of the cardinal, most theoretically important, and most viscerally fascinating aspects of life's history on earth — must be explained largely by the limits and channels of historical constraint, and not by the traditional mapping of organisms upon the clumped and nonrandom distribution of adaptive peaks in our current ecological landscapes. In other words, the inhomogeneous occupation of morphospace largely records the influence of structural rules and regularities emerging “from the inside” of inherited genetic and developmental systems of organisms, and does not only (or even primarily) reflect the action of functional principles realized by the mechanism of natural selection imposed “from the outside.”
In a recent article, Arthur and Farrow (1999, p. 183) pose the key issue in much the same terms: “Why do animal take the forms they do, and not others? Why ... are all land vertebrates 'tetrapods' — except for cases of secondary loss, for example snakes — while none have six, eight, or many legs? Why is the situation precisely reversed for land arthropods? In general, why are certain areas of multicellular morphospace densely populated with many representative species, while other areas, apparently characterizing viable designs, are unoccupied by any extant or extinct animals?” Then, although I would label their distinctions as overly dichotomized and too mutually exclusive (for I seek a fusion of structural and functional influences), Arthur and Farrow also pose the alternatives (1999, p. 183) in much the same manner followed here:
There are two very different answers to these questions, representing two opposing schools of thought on the relative importance of natural selection and developmental constraint in determining the actual distribution of morphologies that we observe ... One is the “pan-selectionist” view that variation is potentially available in all directions from any given phyletic starting-point, and that selection determines which subset [Page 1175] of variants prevails. The alternative is the “developmental constraint” view that many of the gaps we observe between different morphologies do not arise from the non-adaptiveness of the absent forms but rather from the difficulty of making them through an ontogenetic process.
I began this “symphony” of evo-devo with a quotation from one of the great architects of the Modern Synthesis — Mayr's statement, based on adaptationist premises then both reasonable and conventional, that any search for genetic homology between distantly-related animal phyla would be doomed a priori and in theory by selection's controlling power, a mechanism that would surely recycle every nucleotide position (often several times) during so long a period of independent evolution between two lines. The new data of evo-devo have falsified this claim and revised our basic theory to admit a great, and often controlling, power for historical constraints based on conserved developmental patterns coded by the very genetic homologies that Mayr had deemed impossible.
For the sake of both symmetry and logic, it seems fitting to end this section by recalling another quotation by another great architect of the Synthesis, based on the same panadaptationist assumptions about natural selection's controlling power — and also falsified, since then, by new information on historical constraints, impelling renewed respect for formalist themes in revising and expanding our theories of evolutionary mechanisms. But Dobzhansky's closing statement (1951) differs from Mayr's opener (1963) in one crucial way: Mayr's denial of genetic homology represented a sensible consensus for his time; whereas Dobzhansky's assertion of purely adaptational mapping upon ecological places to explain the clumpy population of morphospace made little sense, even at the height of enthusiasm for natural selection's exclusive power — and I can only conclude (as discussed more fully in Chapter 7, pp. 526–528) that Dobzhansky, in his enthusiasm for strict Darwinian theory, had temporarily undervalued a cardinal fact of natural history that his initial training as a systematist had certainly infused into the marrow of his understanding.
In a brilliant opening move, Dobzhansky began the third (1951) edition of his founding document for the Synthesis, Genetics and the Origin of Species, by recognizing the diversity of modern organisms, and the striking discontinuities within this plethora of form, as the central problem of evolutionary biology — at a time when most colleagues would surely have cited modes of continuous transformation, or mechanisms for changes in gene frequencies, within single populations instead. (Despite this unconventionality in subject and level of focus, Dobzhansky opted for a traditional selectionist explanation by titling the first subsection of his book: “diversity and adaptedness.”)
As a wise and wonderful human being, and as a humanist at heart, Dobzhansky began his book with a generous perspective on the meaning and importance of organic diver
sity. The opening paragraph (1951, p. 3) reads: “Man has always been fascinated by the great diversity of organisms which live in the world around him. Many attempts have been made to understand [Page 1176] the meaning of this diversity and the causes that bring it about. To many minds this problem possesses an irresistible aesthetic appeal. Inasmuch as scientific inquiry is a form of aesthetic endeavor, biology owes its existence in part to this appeal.”
After stating the key issue, Dobzhansky then cites the discontinuities within this diversity as the crucial phenomenon demanding explanation. But he begins this second subsection, entitled “discontinuity,” by unconsciously showing his Darwinian commitments in citing organisms as the “prime reality” of biology (whereas, in a hierarchical reformulation of Darwinian theory, several evolutionary levels feature other biological individuals just as interesting, and just as well constituted — with Dobzhansky's beloved species, the quanta of his concern for diversity, as a primary example of individuality at a higher level). Dobzhansky writes (1951, p. 4): “Although individuals [i.e., organisms] limited in existence to only a short interval of time, are the prime reality with which a biologist is confronted, a more intimate acquaintance with the living world discloses a fact almost as striking as the diversity itself. This is the discontinuity of the variation among organisms.”
Dobzhansky then commits his conceptual error in proposing a purely selectionist explanation — externalist at an extreme in its appeal to environmental topography as the sole mapping function for discontinuities in organic diversity — for the crucial fact of dumpiness in the habitation of morphospace. I discussed this passage extensively in Chapter 7 (pp. 526–528), and will only present a summary here. Dobzhansky begins by changing the level of application for Sewall Wright's nonadaptationist model, originally devised to explain why the varied demes of single species may reside upon several discontinuous peaks of an adaptive genetic landscape. By promoting this model to the species level (see Fig. 10-31), and regarding the inhabitants of each peak as a species instead of a deme (and then reconfiguring the peaks as adaptive optima in an ecological terrain, rather than sets of workable genetic combinations among demes, with only the highest peak representing an optimum position for the species), Dobzhansky converted the theoretical meaning of Wright's model from an explanation for why so many demes have not obtained a best possible configuration into a paean for the adaptive optimality of each element in a fauna.
In describing this promoted adaptive landscape, as presented in Figure 10-31, Dobzhansky commits his panadaptationist fallacy by attempting to render the inhomogeneous occupation of morphospace as a simple one-to-one “mapping” of discontinuity upon the external “terrain” that set the selective pressures responsible for crafting all aspects of organic diversity.*
The enormous diversity of organisms may be envisaged as correlated with the immense variety of environments and of ecological niches, which exist on earth. But the variety of ecological niches is not only immense,
[Page 1177]
10-31. Sewall Wright's model of the adaptive landscape “promoted” by Dobzhansky to adaptive peaks for optimal residence of species in an environmental landscape. From the third edition of Dobzhansky's Genetics and the Origin of Species.
it is also discontinuous. One species of insect may feed on, for example, oak leaves, and another species on pine needles; an insect that would require food intermediate between oak and pine would probably starve to death. Hence, the living world is not a formless mass of randomly combining genes and traits, but a great array of families of related gene combinations, which are clustered on a large but finite number of adaptive peaks. Each living species may be thought of as occupying one of the available peaks in the field of gene combinations. The adaptive valleys are deserted and empty.
Furthermore, the adaptive peaks and valleys are not interspersed at random. “Adjacent” adaptive peaks are arranged in groups, which may be likened to mountain ranges in which the separate pinnacles are divided by relatively shallow notches. Thus, the ecological niche occupied by the species “lion” is relatively much closer to those occupied by tiger, puma, and leopard than to those occupied by wolf, coyote, and jackal. The feline adaptive peaks form a group different from the group of the canine “peaks.” But the feline, canine, ursine, musteline, and certain other groups of peaks form together the adaptive “range” of carnivores, which is separated by deep adaptive valleys from the “ranges” of rodents, bats, ungulates, primates, and others. In turn, these “ranges” are again members of the adaptive system of mammals, which are ecologically and biologically segregated, as a group, from the adaptive systems of birds, reptiles, etc. The hierarchic nature of the biological classification reflects the objectively ascertainable discontinuity of adaptive niches, [Page 1178] in other words the discontinuity of ways and means by which organisms that inhabit the world derive their livelihood from the environment.
But the striking discontinuities in morphospace, and their ordering into taxonomic hierarchies, surely don't, at least primarily, “reflect the objectively ascertainable discontinuity of adaptive niches” — and Dobzhansky certainly understood the unstated major reason for such inhomogeneity, even though the strict adaptationism so favored at this time had momentarily clouded his excellent judgment, thus explaining his curious omission. Cats, lions and tigers work admirably well, with each species displaying excellent adaptation to its immediate environment. But the set of all feline species does not clump closely together in morphospace because the summits of an underlying topography now happen to lie at such close mutual proximity in an external world “out there.” Felines form a tight cluster because they share, by historical constraints of ordinary genealogy, a large set of distinctive traits, unique to them alone by virtue of “propinquity of descent,” to cite Darwin's own description of the phenomenon — although each of these traits probably arose for good and conventional adaptationist reasons in a common ancestor. And the larger gap between felines and canines also records, as its primary raison d'etre, a greater separation in history, not the architecture of spacing between two groups of peaks in the current mountain range of worldly ecology.
Sometimes we need the press of new data to inspire the recollection of old truths. I do not doubt that many discontinuities in morphospace represent the colonization by optimal phenotypes of widely dispersed peaks in maximal biomechanical efficiency. But I am equally confident that more of nature's evidently nonrandom, and oddly dispersed, clusters in morphospace, bearing such enormously different weights ranging from single “outliers” to millions of species, primarily record the historical constraints imposed by workable solutions with adaptive origins — developmental designs that then congealed, enforcing reiteration and change within their internally directed channels forever after. Five, for all I know, may be optimal for the radial symmetry of echinoderms, and therefore predictable for any phylum in their domain. But can we argue that the sixfold way of a much, much larger clump marks an optimal and inevitable number for walking, and that elytra represent the only possible design for joint excellence in flight and protection? God, as one of our most celebrated colleagues famously exclaimed, must have an inordinate fondness for these particular creatures if he allowed one design among so many conceivable alternatives to congeal so hard, and then to iterate so often, in nature's wondrous interplay of constraint and adaptation.
[Page 1179]
CHAPTER 11
The Integration of Constraint and
Adaptation (Structure and Function)
in Ontogeny and Phylogeny:
Structural Constraints, Spandrels,
and the Centrality of Exaptation in
Macroevolution
The Timeless Physics of Evolved Function
STRUCTURALISM'S ODD MAN OUTSIDE
In a famous passage from the Introduction to the Origin of Species, Darwin identified the intricately adaptive character of most anatomical features
as the primary phenomenon that any theory of evolutionary mechanisms must explain. Many other sources of information, he states, can easily prove evolution's factuality, but we will not understand the causes of change until we can explain “how the innumerable species inhabiting this world have been modified so as to acquire that perfection of structure and coadaptation which most justly excites our admiration” (1859, p. 3).
Darwin felt that this striking and pervasive functionality of organic design required an explicit functional theory of evolutionary causes rooted in the proposition that adaptive structures originate “for” their utility. As functionalist theories, both Lamarckian soft inheritance and Darwinian natural selection share a defining premise that environmental information about adaptive design somehow passes to organisms, and that organisms then respond by fashioning traits to enhance their competitive ability within these environments. (Above all, functionalist theories require explicit interaction of organism and environment in the service of improving local adaptation. The pure imposition of one side upon the yielding properties of the other side does not qualify.)
The strikingly different mechanisms of the two major functionalist theories — organic response to felt needs for Lamarck, natural selection upon isotropic variation for Darwin — should not obscure their agreement on the key functional principle that adaptation drives evolution as organisms change to [Page 1180] secure better fit to their environments. We prefer Darwin and reject Lamarck because nature's mechanisms of heredity and variation validate the efficacy of natural selection and disprove the existence of soft inheritance, not because we can specify any basic difference in their shared commitment to a functionalist account of evolutionary mechanics.
The Structure of Evolutionary Theory Page 187
