The Princeton Guide to Evolution

III.12

Evolution of Form and Function

Peter C. Wainwright

OUTLINE

1. Form and function in organismal design

2. Measuring the evolution of form and function

3. Key features of life’s functional systems: Multi-functionality, genes, and complexity

4. General principles of the evolution of complex functional systems

GLOSSARY

Adaptation. A process of genetic change in a population whereby, as a result of natural selection, the average state of a character improves with reference to a specific function, or whereby a population is thought to have become better suited to some feature of its environment. Also, an adaptation: a feature that has become prevalent in a population because of a selective advantage conveyed by that feature in the improvement of some function. In this chapter the term is used mostly in the latter sense—as a noun describing a trait that has evolved through this process and helps make the individuals in the population better suited to their habitat.

Fitness. The ability of an individual to survive and reproduce. It can be measured conceptually as the contribution to the gene pool of the next generation.

Functional Morphology. The study of the relationship between form and function in organisms.

Morphospace. Any axis or set of axes that describes the parameters required to illustrate the form of organisms, such as dimensions of parts of the body. By plotting the positions of living forms into a theoretical morphospace it is possible to determine what forms are especially common and which do not exist.

Natural Selection. The process by which organismal traits become more or less common in the population as a function of differential survival and reproduction of the individuals that vary in these traits.

Performance. The ability of an individual to execute tasks important to its daily life. Performance is hierarchical, with fitness being the most integrative, an inclusive measure of performance that sums across many underlying performance traits. Underlying fitness may be running speed, maneuverability, and the ability to avoid detection by not moving (among other performance traits). Underlying running speed are a number of more proximate performance traits, such as the capacity of an individual muscle to generate power, the length of the stride, and the time required for the muscle to relax prior to its next contraction.

1. FORM AND FUNCTION IN ORGANISMAL DESIGN

Form and function are inextricably linked in living creatures because patterns of natural selection reflect the impact of alternative morphologies on the ability of the organism to perform the tasks determining its survival and reproduction. The details of construction of a lizard’s limbs and body will determine how fast it can run and how effectively it can evade specific predators. To the extent that limb dimensions affect maneuverability and sprint speed, they can be expected to evolve if a new sort of predator comes on the scene, favoring a different escape strategy; thus, the primary reason that lizard limb dimensions evolve is that they underlie locomotor abilities, which in turn shape the survival and reproductive success of individual lizards. Form evolves mainly because it shapes performance and hence fitness; therefore, a key to understanding how form shapes fitness is to understand how it determines function.

What aspects of form are most often studied? With mobile animals, most of the body is typically concerned with locomotion, feeding, or reproduction, and these systems are often similar in that they involve muscles and linkages of skeletal elements. The mechanical properties of these systems are shaped to a large degree by the sizes of muscles, the leverage of muscles acting across joints, and the mechanical properties of the muscles and skeletal elements themselves. In vertebrate animals, the skeleton is made mostly of bone; in arthropods the skeleton is made of chitin that is variably mineralized, and in a number of animals there is no skeleton per se, but stiff, contracted muscles often playing a similar role. Approaches to analyzing feeding and locomotor systems are very similar, since the task is to work out the mechanisms by which force and movement are transmitted from the muscles driving the behavior through the system and onto the environment. But not all animal performance is about movement; many other performance traits have a morphological basis. Some examples are camouflage, respiration, attractiveness to potential mates, acuity of vision, and the ability to detect sounds or chemicals. The relevant structures may be microscopic, but sensory systems normally have a performance basis in the size and shape of their component parts.

2. MEASURING THE EVOLUTION OF FORM AND FUNCTION

Studies of the evolution of form and function come mostly in two flavors: population-level analyses in which natural selection is measured, and comparative analyses across species, wherein a longer-term view of the evolutionary process is gained.

Studies of Natural Selection Acting on Variation within a Population

Studies of this type often involve some combination of two approaches in which the impacts of form on performance and on fitness are separately estimated. The strength and form of natural selection can be measured on size or shape of structures, using the standard methods of studying natural selection in natural populations (see chapters III.3 and III.4). The typical strategy in this type of study is to capture a large number of individuals in a population, measure the traits of interest in each individual, and mark the individuals with an identifying tag before returning them to their habitat. After sufficient time for some mortality or growth to occur, individuals are recaptured and their identities recorded. The starting form of survivors is compared to the entire original sample to determine whether form affected (or is correlated with) the probability of survival. In a handful of studies, researchers have measured not only form but also performance traits after initially capturing individuals, permitting deeper understanding of the ways in which selection is acting on performance and its underlying traits. In most cases, however, researchers measure selection acting on form without knowing the relationship between form and performance. In these cases they either assume they know how form and function are related, or they simply are not interested.

Understanding the functional significance of form deepens our understanding of adaptation. A standard approach here is to model the functional system, such as the wing or bill of a bird, then to parameterize the model with key measurements from animals to estimate functional properties expected to directly impact performance. Here, measurements of individual size and shape are used to estimate a functional property. A common example with muscle-skeleton systems is to measure the mechanical advantage of a muscle acting across a joint with an input lever and an output lever. This might be done in a jaw by measuring the distance between the attachment of the jaw muscle on the jaw to the jaw joint as the input lever, and the distance from jaw joint to the location on the teeth where the food is held as the output lever. The ratio of the input lever to the output lever gives the mechanical advantage—or the proportion of an input force (produced by the jaw muscle) transmitted through the lever to the food item. Mechanical advantage of levers reflects a trade-off between transmission of force and transmission of movement (usually the movement of a contracting muscle). A lever with a low mechanical advantage transmits less force, but more motion, than a lever with a high mechanical advantage. In fact, there is a one-to-one exchange of force transmission and movement transmission in a lever, so that mechanical advantage modified during evolution to enhance force transmission will transmit less movement. This model is based on principles from physics. Levers are central to mechanical engineering, but other functional systems might be more appropriately modeled by using chemical engineering, optics, or electrical engineering. Whether the system is a lever or some other functional device, the modeling approach is especially good for comparative studies because it allows one to efficiently compare the functional implications of variation in form. The use of models has also been very useful in studies of large numbers of species to relate form and function to habitat use and feeding habits of animals.

Comparative Analyses

The modeling approach is a powerful way to study diversity across species. In this arena an important additional tool is an estimate of the evolutionary relationships among the species being studied, or a phylogeny (see chapter II.1). The phylogeny provides a road map to the sequence and pattern of evolutionary modification of functional systems and can be used to identify the sequence of assembly of complex adaptations, as well as patterns of association between form and aspects of the species’ ecological niches, such as habitat and feeding habits. By focusing on variation in parameters from functional models, one can draw a tighter, more causal link between form and these ecological patterns.

In a comparative analysis, one infers a past process by evaluating the outcomes of evolution. By studying numerous evolutionarily independent transitions to the same habitat or feeding specialization, one can develop an understanding of whether the modifications to form accompanying the transition always occur in the same way, or alternatively whether multiple solutions to the switch have occurred during evolution. The combination of species values of traits and the phylogeny allows an estimate of sequential changes that have occurred. This is particularly useful if one is interested in how a complex novelty came about during evolution.

One of the classic questions in the evolution of organismal designs concerns whether complex functional systems are rapidly assembled during evolution or are assembled piecemeal over a long period of time. One example of this sort of analysis is the origin of powered flight in birds. Did all the underlying innovations for powered flight occur at one time in the ancestor of birds? The phylogeny of birds and the closely related theropod dinosaurs from which they arose reveals a fascinating story. Both bipedalism (walking and running on two feet) and feathers clearly evolved long before the origin of powered flight. In fact, the surprisingly ancient history of feathers within theropods indicates that feathers, an integral and crucial element of powered flight in birds, most likely evolved for their insulating value and were only incorporated into the suite of adaptations for flight after many millions of years of keeping theropods warm. The evolution of powered flight involved the modification of feathers, both as key structures in the wings and to help produce a smooth body surface facilitating efficient airflow. In addition to modifications of feathers, the muscles and skeleton of the forelimb were modified considerably into a structure that supports the flight feathers and moves powerfully in the pattern used during flight. This is one of many examples in which the phylogenetic context of complex adaptations provides surprising and interesting insights into their evolutionary history.

Perhaps not surprisingly, the dominant pattern that has emerged in studies of the evolution of complex innovations is that key components of the system evolve earlier and are modified for a new function as the system is gradually assembled. The importance of this common pattern is that it underscores the dependence of evolution on available building blocks. Some major, complex innovations have huge effects on the subsequent evolution of the lineages that possess them. Innovations such as powered flight, the origin of endothermy, and air breathing in fishes are major breakthroughs that opened up whole new ways of life. Some major innovations are followed by a burst of diversification as organisms radiated out into the new niches made possible by the innovation (see chapter VI.10). Interestingly, it is much more common for even the best innovations to show little impact on the success of the group for a prolonged period of time, and in some cases there is never a spectacular burst of diversification. This last point is illustrated by a major innovation in the feeding system of fishes called pharyngognathy, in which bones and muscles associated with the gill arches are modified into a functionally versatile second set of jaws. All bony fish have this second set of jaws, but in six lineages of ray-finned fishes this system has been independently modified in a similar way, making the jaws powerful and very versatile for processing food items. This condition, pharyngognathy, evolved independently in cichlids and in labrids (wrasses and parrot fish), two of the most successful and ecologically diverse groups of fishes in tropical freshwater systems and coral reefs, respectively. In these groups, pharyngognathy seems an important innovation that contributed to exceptional diversification; however, pharyngognathy also evolved independently in damselfishes, halfbeaks, surfperches, and false scorpion fishes, but these groups have thus far failed to diversify ecologically to any notable degree. The point is that innovations, in and of themselves, do not guarantee spectacular diversification. Lineages possessing valuable innovations must also find themselves in appropriate ecological circumstances that promote realization of the potential provided by the innovation.

3. KEY FEATURES OF LIFE’S FUNCTIONAL SYSTEMS: MULTIFUNCTIONALITY, GENES, AND COMPLEXITY

So, on the surface of the problem it is fairly straightforward to evaluate the relationship between form and function and the adaptive value of size and shape of organisms and their parts—by determining which designs confer higher performance, and which have higher fitness. But several factors, other than the direct linkages between form, performance, and fitness, characterize biological systems and make the study of the evolution of form and function especially interesting.

The first is that functional systems in organisms do not occur in isolation. Most structures participate in multiple performance traits that all compete for the shape and properties of the structure. You might think the wings of an insect would be shaped solely for their role in powering flight, a physically demanding function, but insect wings can also be used to absorb warmth that radiates from the sun and to provide camouflage when the animal is at rest. Ultimately, the wing size and shape evolve that maximize fitness, or the integration of all the underlying performance traits that directly or indirectly determine reproductive output. Because the ideal wing design may differ for different performance traits, wing form reflects trade-offs that balance these different demands. In some cases other functions of insect wings have become so important that the animal has lost the ability to fly. An example is an eastern North American katydid, Pterophylla camellifolia, in which the forewings of the male are inflated to form a resonating chamber used in sound production, and the animal cannot fly.

Trade-offs come about when a single structural trait contributes to two or more performance traits. They are thought to be one of the dominant factors explaining the diversity of organismal design because of the inherent constraints they place on maximizing performance. The fact that most parts of an organism participate in more than one performance trait is a major reason that systems can be honed by natural selection but not reach the highest possible performance in any one functional system. It is often possible to evaluate the adequacy of design in a particular system for a particular function, but it may be difficult to identify all the performance trade-offs the organism faces in building these structures.

A second factor is that organismal form is a consequence of genetic programs that normally allow single gene products to contribute to the formation of many structures (i.e., the genes are pleiotropic). As a result, we can expect it to be difficult to modify genes to effect specific morphological changes without causing other changes. These genetic correlations can limit adaptation, at least in the short term, because the unintended, correlated changes to form may have a negative consequence for other performance traits. This constraint is alleviated to a considerable degree by trait-specific differences in gene regulation in different parts of the body, but the potential for manifold impacts of changes to widely used genes is a major factor in the evolution of developmental pathways.

A third factor is complexity itself. Perhaps no other intrinsic feature of living functional systems has so much impact on the dynamics of their evolution as the fact that they are inherently complex. Even at its most basic levels, complexity impacts the evolution of functional systems. Potential diversity is fundamentally a function of the number of parameters required to describe a form, or its degree of complexity. In general, the story of the ways in which intrinsic properties of functional systems influence their diversification is a story about the many implications of complexity.

4. GENERAL PRINCIPLES OF THE EVOLUTION OF COMPLEX FUNCTIONAL SYSTEMS

The nature of the relationship between form and function plays a prominent role in shaping evolutionary dynamics of functional systems, and a number of processes have been identified that highlight this interplay. It is not simply the adaptive value of traits that govern their evolution; the way in which they impact function is also important. Let us consider, for example, how trade-offs can change or even be abolished during evolution.

Decoupling Trade-Offs

We observed earlier that structures in organisms normally participate in multiple performance traits and that these competing demands can lead to trade-offs limiting diversification. One key to overcoming this sort of constraint is to break the linkage between the two performance traits. If the structure in question must no longer serve both performance traits, the constraint would be released, perhaps allowing modifications of the structure to underlie the evolution of an enhanced ability in its function. There are many examples of this sort of decoupling of performance traits during evolution. One involves the feeding system of fishes. Primitively, fishes (like most other vertebrates) capture and process prey with their oral jaws. This is the condition seen in cartilaginous fishes and a few lineages of bony fishes, such as sturgeon and lungfish. But prey capture in an aquatic environment, without the benefit of appendages, is dominated by the highly dense and viscous nature of water. As a result, the vast majority of fishes use suction feeding to capture prey. In suction feeding, the mouth is rapidly expanded, drawing in water and prey. Effective suction feeding on quick elusive prey is enhanced by light bones and levers in the jaw that favor the transmission of displacement over force. But once caught, prey is processed by biting; in particular, if the prey has a tough outer shell, robust jaws are needed with levers favoring force transmission over movement. So when the oral jaws are used for both prey capture and processing, trade-offs can limit adaptation in both functions and overall diversification of jaw design.

This trade-off was broken with the origin of teleost fishes, when a second set of jaws evolved, the pharyngeal jaws. These new jaws are located at the back of the oral cavity just in front of the esophagus; following their origin, they took over the role of prey processing, releasing this function from the oral jaws and permitting much more extreme oral specializations for prey capture. The introduction of a new set of jaws that came to specialize in prey processing released a major constraint on the oral jaws.

This is an example of what is called a functional duplication: when a novelty arises that can perform a function of an existing structure. The general implication is that performance of the function by the new system releases constraints on the form of the original system. The ultimate consequences of duplication for the original system may be to enhance other performance traits, or even to permit breakthroughs in design that had been prevented by the need to perform both functions. Functional duplication has the potential to increase the overall performance of the organism and is a common evolutionary phenomenon operating at many levels of biological design. For example, gene duplication is thought to be one of the most common mechanisms of gene diversification and is the focus of a huge amount of research. The principles operating in gene duplication are effectively the same as those operating at the level of the organism.

One way two performance traits can become decoupled is for one to be taken up by a second system, through evolutionary modifications. A second way is when the original structural system is duplicated, and a descendant copy is later modified to specialize in a different performance trait. Exactly this history has been rather common among certain body parts and groups of organisms. Decapod crustaceans have body plans based on repeated body segments with homologous parts. Appendages and their function vary from one part of the body to another, and species can differ considerably in the degree of diversity in appendage function along the body. Some of the diversity includes use of appendages as thrust devices in swimming, legs for walking, claws for defense and prey processing, and mouthparts for chewing. A less well-appreciated phenomenon occurs in muscular systems, when muscles become subdivided into two descendant muscles. Typically the two new muscles attach to the same structures as the original muscle, but over the course of time, one may migrate and develop novel attachments and novel functions. This phenomenon is important in the evolutionary history of tetraodontiform fishes—puffer fishes, triggerfishes, and their relatives. Primitively, these fish have two muscles that attach to the jaw and provide the power behind the bite. These muscles have been subdivided numerous times during evolution so that some lineages have five, eight, or even ten jaw muscles. Across tetraodontiforms several examples can be found in which the muscle no longer attaches on the jaw and has evolved an entirely novel function in compressing the oral cavity.

To summarize this section: alterations of form occur by which performance traits are decoupled and constraints on morphology alleviated. This is one major way in which complex organismal systems evolve.

Many-to-One Mapping of Form to Function

Another mechanism exists by which complexity enhances the evolutionary potential of organisms. When performance and functional traits have a complex underlying basis, almost always, many combinations of those underlying parts will give any particular value of the functional trait. As an example, consider the oral jaw biting system of a vertebrate in which the strength of the bite is a function of just two traits: the force-producing capacity of the jaw muscle multiplied by the mechanical advantage of the lever through which it acts on the jaw. Many combinations of muscle force and mechanical advantage will give any particular value of bite strength. This many-to-one mapping of form to function is inherent in any performance trait determined by multiple traits.

Many-to-one mapping of form to function has far-reaching implications for the evolution of functional systems. In general, this phenomenon acts to soften the impact of trade-offs, because there is almost always more than one way to modify an existing system to create the functional properties favored by natural selection. There are at least three important evolutionary consequences of many-to-one mapping of form to function. First, and most important, the capacity for multiple forms to have the same functional property can permit the optimization of two or more functional properties shaped by the same structures, even when changes in some components result in a functional trade-off. This highly nonintuitive result comes about because of the potential for alternative values of underlying parameters to produce a single value of one function, while permitting change in the second function. In essence, changes to dimensions of structures may be neutral with respect to change for one function, while producing a potentially adaptive change in a second function. In cases where this phenomenon is looked for, it has virtually always been found. The flexibility conferred by many-to-one mapping on design releases constraints on evolution by providing pathways through morphospace with little or no effect on key functional properties of the organism.

The second consequence of many-to-one mapping is that lack of a one-to-one match of form to functional properties means that there is the potential that morphological and functional diversity will be decoupled. Imagine a group of 12 species of finches that feed on prey ranging from insects to seeds of various sizes and hardness. If we measure the dimensions of the bill and its levers as well as the jaw muscle used to close the bill, we can calculate the diversity among the species in these traits. One common way to measure diversity among species in trait values is to calculate the variance among species in the trait. Total diversity of several traits can be obtained by summing the variances of several traits. We can also measure diversity among species in the capacity to exert a forceful bite. Because many combinations of bill-lever mechanical advantage and strength of jaw muscle can produce a given value of bite force, it is very likely that diversity among species in the morphology will be greater than diversity in bite force. The potentially tight relationship between morphological and functional diversity is weakened by many-to-one mapping. Conformation at the lower level precisely determines functional properties at the level above, but as we have seen, the reverse is not true when many forms have the same functional property. Lack of correlation between diversity among species in form and diversity of functional properties is surprisingly common, and just one of the reasons that knowledge of how an organism works is necessary before the meaning of variation in structures can be interpreted. Some variation is neutral with respect to function, but these emergent properties of complexity make it possible for decoupled diversity even when the form directly underlies the function.

The third consequence is that many-to-one mapping results in a strong signal of evolutionary history in the design of organisms. If only a single combination of muscle force and lever mechanical advantage produces a given value of biting force, then when natural selection favors jaws with this biting strength, it will always produce the same form. But strict convergent evolution (the independent evolution of the same form in response to the same selective pressure in two lineages) is actually surprisingly uncommon. Different lineages typically start at different places in morphospace; when natural selection favors a specific functional property in different lineages, it is more likely that the ancestral form will simply move to the closest form satisfying the function, rather than always converging on the same combination of traits. Natural selection will produce a combination of traits giving the optimal functional property, but not a particular one rather than other possible combinations, unless they have negative consequences for other functional properties. What this means is that the form produced during a bout of evolution in response to a specific selective force will depend in part on the starting form in the population when the response began. In other words, there will be a strong phylogenetic signal during the evolution of morphological systems.