Improbable Destinies: Fate, Chance, and the Future of Evolution II

The bulk of the book is devoted to describing experiments that have studied evolutionary changes in specific species over short periods of time. Ever since it was noticed that peppered moths, which before the industrial revolution had an appearance that allowed them to blend in with tree bark, had become dark-colored and blended in with sooty backgrounds, there has been scientific interest in the timeframe of evolutionary change. Several of the experiments included in the book study the effects of predators on the physical characteristics of their prey. When predators were introduced to the habitats of light-colored guppies that stood out against a dark background, the guppies soon became dark-colored. When predators were introduced to the habitats of Anolis lizards that lived near the ground, the lizards quickly evolved shorter legs, which permitted them to live higher up in trees. Other environmental changes also induced adaptations. Grasses that were exposed to different soil conditions tended to evolve different characteristics according to their specific soil composition. E. Coli made rapid biological adaptations according to the food source available. There are also descriptions of experiments with stickleback fish, deer mice, fruit flies, yeast and other organisms.

Losos's discussion is loosely framed around the differing views of evolution presented by Stephen Jay Gould and Simon Conway Morris. Gould is taken to emphasize chance as an intrinsic element, whereas Conway Morris is taken to emphasize the likelihood of convergence in similar environments. Gould seems to be saying that a one-time change in the distant past may fix certain aspects of the course for all descendant organisms, whereas Conway Morris seems to be saying that organisms from different lineages may be shaped by the environment to become similar organisms. The experiments seem to show that both are partly right, but that convergence is not a universal phenomenon. Gould's views are often confused by his use of stylistic flourishes in Wonderful Life and other books. All of the experiments are limited by technical and conceptual problems. Experiments done in the wild lack controls, since the exact composition of each ecosystem isn't known or replicable. Even in more controlled lab settings, it is possible that, for example, unintended minor variations in vial temperatures produce different results.

The conceptual problems are more serious. From a scientific standpoint, there is always the stipulation that correlation does not imply causation. Thus, particularly in the field experiments, the actual physical causes behind each biological change may not be known. In a lab experiment in which there is a unicellular organism of known genetic composition, causation may be easier to determine, but there are still enough variables to make that difficult. Another basic problem has to do with the repeatability and predictability of an experiment; in biology this is far more problematic than in physics or chemistry. Finally, specific to biology is the question of phenotypic plasticity, or an organism's range of physical variability within its species. For example, a chameleon has the ability to change its color without becoming a different species, while most other organisms do not. Without an intimate knowledge of an organism's genetic makeup, including the genetic basis of its specific phenotype, it may be impossible to know whether an immediate environmental change has caused an evolutionary adaptation, as opposed to a variation within an existing genome.

Losos uses New Zealand as an example against convergence, since it contains no indigenous mammals or animals similar to mammals, though mammals flourish in other parts of the world in similar environments. Thus, the environment doesn't necessarily cause specific life forms to evolve. The experiments seem to show that something resembling convergence may occur when genetically similar organisms are placed in similar environments. However, organisms with significantly different evolutionary histories seem unlikely to respond similarly to the same environmental pressures. Therefore, as a general thesis, Conway Morris's version of convergence seems incorrect.

On a theoretical level, I have been thinking about how the very concept of species may itself be a man-made idea that simplifies the world for us but actually has less applicability than we think it does. Species that reproduce sexually have somewhat arbitrarily been defined as organisms in which the males and females produce non-sterile offspring. Sexual reproduction itself is a primary vehicle of evolutionary change, because parents do not have identical genomes, and one parent may confer a genetic advantage to offspring that the other does not. This aspect of evolution is hardly discussed in the book, because it would be more difficult to test experimentally than the purely environmental tests included. Large organisms such as humans can be seen as symbiotic collections of trillions of microbes. If humans were to become extinct, many of the species of microbes in our bodies would survive and continue to evolve. In that case, one might argue that environmental pressures arose and only the best-adapted microbes survived: perhaps this was merely a habitat change for the microbes. At this level of evolution, humans can be seen as a sort of meta-organism or superorganism made up of symbiotic microbes. Although we have many legitimate reasons to think of ourselves as a species, the fact is that, with the exception of identical twins, every person is different from every other. When you get down to the nitty-gritty of evolution, the processes are so complex that our categories seem inadequate. Ultimately, I think that determinism is at work, but in such a complex manner that we are unable to comprehend it at the subatomic level, and that we cover this up by using terms such as "species" and "randomness," though they can't really do the job.

Overall, I found the book interesting and informative, but its use of conventional publishing gimmicks to avoid scaring off sciencephobes doesn't really change the fact that it is probably of greater interest to evolutionary biologists than to the general reading public. Significantly, according to Losos, evolutionary biologists sometimes have to avoid using the e-word so as not to offend people whom they encounter while conducting their research. One would have hoped that by now science would have come out of the closet.