built. Cell division provided the mechanism for reproduction of the cell as an organism. Multicellular organisms are aggregates of differentiated cells that ultimately descend from a single cell (e.g., the fertilized egg). Thus, large complex organisms are built on a process of duplication with variation.

Organisms are modular in many ways beyond being aggregates of differentiated cells. Many if not most higher-level structures, like organ systems, are also modular (each themselves built up of cells, of course). A limited number of basic processes seem to be responsible for this hierarchical modularity, which we will review in later chapters. These processes are responsible for initiating very local cellular division and differentiation to produce individual organ subunits like leaves, flowers, intestinal villi, feathers, teeth, nephrons in kidneys, ommatidia in insect eyes, or vertebrae. Somewhat similar interactions may be responsible for the branching, a related but somewhat different process, that produces repetitive pattern in plants, lungs, blood vessels, and other structures.

A duplication strategy applies to physiological as well as morphological systems. The lipid (fat molecule) transport, endocrine (hormone), and immune systems, for example, are characterized by the interaction of slightly different products of related genes (the modules). How these various types of modularity arose and work will be discussed throughout this book.

Duplication and modularity were of course long known to be important in life but were not widely considered to be basic properties of evolution, because the phenomenon per se can occur without them. But early thinkers in the history of modern biology, for whom evolution meant the development of an organism, considered it; Goethe even likened the repetition of bones (as in vertebrae) to that in plants (leaves) (Richards 1992). The founders of biology were groping in the 19th century for such generalizations, or laws, of biological traits (at that time, essentially meaning morphology, because that was mainly what could be studied) and its variation, and for example how to explain the morphological changes through which embryos go, or the morphological or embryological similarities or differences (divergence) among species. Darwin used many of the same facts, but he accounted for them in terms of historical rather than developmental time. His idea provided a sweeping generalization about external processes that produce organismal diversity. After his and Wallace's exciting new idea, the central role of development, and the search for comparable internal generalizations, were somewhat shunted aside (e.g., Arthur 2002). They have returned.


Life as we know it depends on both precision and chance. The elegant precision of DNA replication gives life its predictability, generation after generation, but without chance mutations in genes, as well as the stochastic and error-prone nature of cellular processes in general, evolution would have no variation with which to work. We would all still be swimming in the primordial soup.

Darwin and Wallace had flashes of insight when they thought of the struggle for existence occasioned by overpopulation and the idea that the fittest organisms would prevail. This is the origin of the notion of natural selection, and both Darwin and Wallace saw selection as an immensely powerful tool for modifying organisms. To Wallace, and to some extent Darwin, the assumption was that the organized aspect of life is the product of selection, but in many ways this is as much a belief system for biologists as is that of fundamentalists who view the world of organisms as having been directly created by God. There are important reasons why we need to be careful about this view. One is our own human- and culture-dependent view of what it means to be "organized."

The second is a form of the anthropic principle, in essence, that things we see are not as improbable as they may seem, because had they not existed neither they, nor we, would be here to see them. Even in purely darwinian terms, whatever is here had to be "adaptive" in the sense of having been reproductively successful. Adaptations may seem highly refined, but among all the essentially infinite number of ways organisms might have evolved (even if by chance), something evolved. Thus, whatever is here is not as unlikely as we might otherwise think.

This of course cannot be used as an argument against natural selection; it should merely temper our after-the-fact reconstructions. However, the other side of causation is chance, and chance is pervasive and unavoidable—and often, as we will see, almost indistinguishable from causation. Darwin and Wallace were impressed by the universal ability of organisms to produce more offspring than could in the long run survive. Often, this means massive die-off. No matter how many acorns an oak tree produces or how many eggs a fly lays, most of the time populations are relatively stable in total size. Even a very small growth rate leads to major population size changes in short numbers of generations (a property of exponential growth). Thus, an oak tree on average produces one descendant tree.

This massive die-off means that differential survival based on the characteristics of the particular lucky acorn may over time favor that variant on the oak, but it doesn't imply that it will do so. In fact, it seems clear that life on Earth is much more awash in the disorder of chance than most biologists tend to acknowledge. At least, if the world is as remorselessly ruthless as the usual darwinian view holds, it is wanton rapine, directed to no particular end.

Actually, evolutionary geneticists recognize that much—perhaps most—of the genetic variation in the world not only arose by chance mutation but probably is barely if at all affected by natural selection: the amount and nature of variation changes over time by chance aspects of survival and reproduction alone. The same seems as likely of traits as it is of genes, as we will discuss in Chapter 3.

But What Is "Chance"?

Formal discussion of chance is a profound subject in cosmology about which there is no consensus; in this book we are not really concerned with whether true chance events occur in nature or whether they follow regular textbook probability distributions (e.g., binomial, normal, etc.). Instead, by "chance" we mean literal, or at least practical, unpredictability.

A probability distribution essentially provides a formula by which we can compute the relative likelihood of a given outcome in an experiment that can be repeated, like coin tossing. Sometimes observations can't be repeated in practice, but we still assume that underlying what we can actually observe is such a distribution. The Brownian motion of molecules is an example. But what about the "chance" that a given enzyme molecule meets a molecule of its substrate in a given cell? This may be a case in which we can at least assume that this probability could be specified in principle. In practice, this is not so clear: we can't really come close to specifying the probability because a cell is not a uniform fluid space, and there is no such thing as an archetypal cell in the practical sense, except perhaps in the somewhat artificial conditions of an experimental lab. Genes are expressed in cells based on regulatory mechanisms that, while specific, entail a substantial component of chance, but one that would be very difficult to predict. Each cell of a tissue of seemingly identical cells is somewhat different. What about the chance that a given wildebeest will be eaten by a lion or that a given human will have a given number of children during her lifetime? These questions seem somewhat simpler and more meaningful, and we can imagine replication or sampling distributions that could help us answer them, in principle. Many aspects of life and evolution, however, involve chance in a more profound sense. Biologists might try to estimate the probability that frogs could have evolved "by chance," for example, a question that is analogous to asking whether, if we could start life on Earth over again we would find the same outcome.

This is really a colloquial but scientifically misleading misuse of the term "chance" because there is no seriously meaningful sense in which the outcome of life represents a probability distribution in the sense of repeatable experiments. At best, the number of possible outcomes is so complex that it does not make real probabilistic sense to speak of the "chance" that frogs will evolve in the most stringent use of the criteria of science. At worst, our understanding is so rudimentary that the question itself makes too many assumptions that cannot be verified. In the end, we cannot operationalize this statement by specifying any practical way to make it testable. It is a metaphor.

Much of life is this kind of metaphor. It is important to realize that, although chance plays a role throughout life and its evolution, much of the time what that means essentially is unpredictability for all practical purposes. Some of the time— an unknown amount of the time—it means literal, ultimate true unpredictability as far as what we understand of the world today can tell us.

Implications of Sequestration and Modularity

Communication, differentiation, nesting and repetition are seen throughout life. They depend on controlled degrees of sequestration. In multicellular organisms, cells are organized into tissues of different types, then into organs or organ systems, such as the vertebrate epithelial, nerve, muscle, blood, lymphoid, and connective tissues, or, in plants the dermal, vascular, and ground tissues. Individual cells are bound together in layers in various ways and need modes of both communication and adhesion (and, for some cell types, such as sperm or egg or circulating blood cells, means of nonadhesion). Intercellular contact and communication allow the "blending" of isolated material to a limited or controlled extent.

The sequestration among cells in an organism is less than that between individuals in a species or between species. One of the most important of the mechanisms for reducing isolation of individuals within a species is sexual reproduction. Recombination among homologous chromosomes during meiosis breaks down some of the isolation of individual genes, and recombination of genomes (the entire set of genes inherited by an organism) in the formation of diploid zygotes allows diversity that would otherwise accumulate separately to be mixed, presenting greater variation and ability to respond to environmental circumstances among other things. Sexual reproduction allows individuals to be different and, for example, subject to screening by natural selection but also allows a limited and highly controlled amount of exchange that means that species can exist.

SOME BASIC ASPECTS OF THE CHEMICAL "LOGIC" OF life Here it is worth mentioning two facts that are fundamental to the way life works from a chemical point of view. Many other generalities apply, such as the properties of carbon-based life with the other major molecules (hydrogen, nitrogen, and oxygen). But here we are not referring to basic biochemistry but to the logic of evolution.

The details will be seen throughout this book. The first basic principle of the logic of life is that the four nucleotide bases, commonly denoted A, C, G, and T, chemically pair: A with T and C with G. As explained in Chapter 4, this is the essential fact in the "information" storage property of life.

The second major fact is that proteins, the other basic functional constituents of life and evolution, function by combining with other chemicals in the cell. From an evolutionary point of view, this in particular includes interactions of proteins with each other. This is fundamental to communication of all sorts, to the basic biochemical aspects of life, and to the way that evolutionary history and information are stored and used. An important example is the general phenomenon of one protein (known as a ligand) binding with another (known as a receptor) that has been evolved specifically to chemically recognize the ligand. We will see this phenomenon throughout the book.


In the context of adaptation, sequestration, and the various levels of hierarchy seen across life from intracellular reactions to species evolution, it is important to keep in mind that these processes are contingent. That is, what happens now depends on the current state and not on previous or future states. The myriad biochemical reactions taking place within a cell are often localized within the cell, and each reaction depends on the current state, even if the raw materials of earlier states are still around or those prior-stage reactions are still occurring.

Two major points about this are worth raising in this context. First, life today depends on its history of evolution, and in that sense the nature of a cell or organism is dependent on its history as "written" in the genome (and in the nature of the environment around it, also the product of evolution). But what happens today is contingent on today's state, rather than some previous one. And any apparent longterm trajectory (e.g., teleology, or changes aimed at a certain distant adaptive end) is illusory except to the extent that the nature of things that are interacting today make it possible to predict what will happen tomorrow.

Beyond "tomorrow," it is generally agreed that life and its evolution are not specifically predictable, and the reason is contingency: what happens in the future depends on events that we view as purely chance, such as mutation, climate change, and the proverbial "acts of God" such as being struck by lightning. Similarly, it may be that many systems in nature are "chaotic," meaning that, even if they are totally deterministic, only perfect knowledge would allow us to predict future states with specifiable accuracy (a hypothetical example is the well-known one of a butterfly flapping its wings, unseen and unmeasured, that eventually leads to major weather changes). Whether such systems exist is itself essentially unprovable, but if it were to be shown that there really is nothing in nature that is purely chance—that is, that all the "laws" of nature are perfectly deterministic—would we have to back away from the view of nature as contingent? If so, everything really would be predictable from the beginning, a form of omniscience that would be truly God-like.

There are traits in some species that do not seem to vary, yet the species can be shown to harbor unexpressed variation (e.g., Dun and Fraser 1959; Gottlieb et al. 2002; Lauter and Doebley 2002) that can be revealed by changes in circumstances or other means—a kind of potential that is already there in the organism. Some aspects of life do not seem as chemically unlikely as previously thought. Functional proteins have been shown to arise rather easily even in random mixes of amino acid strings (Keefe and Szostak 2001), that is, they can be a kind of natural state.

Except for the stochastic element affecting whether two molecules come in contact, it seems generally true that the nature of molecular interactions is built into their atomic structure, that is, the principles of basic chemistry. Chemistry determines how DNA does its complementary base-pairing, the specific interactions of proteins or how enzymes interact with substrate molecules to affect chemical reactions. If there were no stochastic factor affecting whether molecules come into contact, the genome and other molecules within a cell would have essentialistic

Was this article helpful?

0 0

Post a comment