Ford Denison . Darwinian Agriculture. How Understanding Evolution can Improve Agriculture . Princeton University Press . Princeton and Oxford 2012 . 258 Pages .

Although there is little doubt that humans were inspired by birds when they developed aircraft, it is usually less well appreciated that aircraft also increased our understanding of bird's flight. The artificial recreation of a biological system may thus improve our understanding of its functioning. Something similar has occurred with agriculture, albeit on a longer time scale. The start of human agriculture some 10,000 years before present, ultimately offered Darwin a rich source of inspiration for developing his ideas on natural selection, reflected in the title of the first chapter, “Variation under domestication,” of his masterpiece. He found his best argument for the power of natural selection in the artificial selection of domesticated organisms by humans. Surprisingly, the opposite, the application of evolutionary ideas to agriculture, has not been exercised until recently. Ford Denison now fills this gap with a book entitled Darwinian Agriculture.

Denison sees three core principles of Darwinian agriculture that should lead to productive, efficient, stable, and sustainable agriculture. First, prolonged natural selection has rarely missed simple, trade-off free improvements. Simple improvements by genetic engineering are therefore highly unlikely. Second, competitive testing is more rigorous than testing merely by persistence. Ecosystems may be stable, but their stability is not a result of natural selection among ecosystems. Ideas from ecosystem patterns thus require more rigorous testing than ideas inspired by the competitively tested individual adaptations of wild species. Third, we should hedge our bets with a greater variety of crops and ideas. These principles are worked out in 12 chapters. The book sets out to formulate a research agenda for yet another application of evolutionary theory after the now well-established research field of Darwinian medicine. Similar to Darwinian medicine, Darwinian agriculture is broad, ranging from applying evolutionary insights to efficient disease control, artificial selection of complex traits, artificial group selection, and selection for better mutualists. The examples Denison uses reflect this diversity.

Throughout the book, Denison balances between two extreme views on agriculture, which he both rejects. The one view is that biotechnology will solve all current and future problems, and improve agriculture by simple means. Denison argues that it is highly unlikely that millions of years of natural selection have missed simple, trade-off free opportunities. Therefore, “trade-off blind biotechnology” will rarely work, at least not by simple tricks. For example, increases in drought tolerance usually will lead to reduced photosynthesis. A thorough review of the literature shows that neither yield potential, photosynthetic efficiency, water use efficiency, or nutrient use efficiency have increased very much recently, despite the promises of biotechnology.

Denison identifies two classes of genetic improvement, where he “does” see a role for biotechnology. First, while it is unlikely that “simple” solutions have been missed by natural selection, this may not be true for “complex” solutions. The reason is that such solutions will require multiple genetic changes, some of which individually may decrease fitness. Therefore, to achieve such solutions via natural selection would require the simultaneous occurrence of multiple mutations, which is highly unlikely. Evolution thus may have stopped at a suboptimal solution before a better solution was found. Although, in principle, biotechnology can be useful to achieve such complex changes, these changes will also be the most challenging to achieve via biotechnology and will require a thorough understanding of the biology of the organism. The failure so far to transfer C4 photosynthesis from corn into rice illustrates this. Second, if we understand the trade-offs and can create the conditions where these trade-offs are tolerable, we can find solutions, which were weeded out by natural selection in the past. The most important category of these trade-offs is that between individual fitness and group productivity (see below).

The other extreme view rejected by Denison is that agricultural ecosystems should resemble natural ones as closely as possible. The main criticism against this standpoint is that usually individuals, and not groups, let alone complete ecosystems, have been tested by natural selection. However, Denison leaves open the possibility that some aspects of the organization of natural ecosystems make them good models for agriculture, but the evaluation of these aspects usually requires thorough testing. For example, mixed cultivation may be an important means to limit the opportunities of specialized pathogens, by hindering their dispersal. Denison critically reviews the evidence for this popular idea and considers various ways of mixed cropping both in time and space and at different scales. A nice analogy is provided by a phenomenon called “mast seeding” found in some trees, with low versus high seed production in alternate years, often synchronized among trees. This alternation can be an effective defense against seed-eating pests, which will be kept at low frequency during nonproductive years, and cannot increase sufficiently fast in population size to eat all seeds in productive years. Also some periodical insects employ this strategy, with cicadas with a 13- or 17-year periodic cycle as the best known example. A lesson from these natural examples, Denison argues, is that crop rotation with diversity in time, rather than in space, may work.

So usually individuals, and not groups, have been tested by natural selection. Mast seeding and periodical reproduction in insects are strategies driven by individual selection that also benefit whole populations. Often, however, adaptations maximizing individual fitness are harmful for the group. Here, Denison has identified an important category of possible applications in agriculture, to a large degree missed so far. In contrast to natural conditions, which are usually not favorable for between-group selection, we may artificially create these conditions. Some of these opportunities have already been exploited, be it unconsciously. For example, natural selection acting on individual plants competing for light has favored increased stem length. It turns out that this individual adaptation “decreases” group fitness. In fact, the main increase in yield in wheat during the so-called green revolution is due to a shift in investment from stem toward grains (and a reduction in wind damage to crops). Similarly, gardeners nowadays grow their trees as low-stemmed rather than high-stemmed trees, knowing that this maximizes yield per hectare.

These examples illustrate the importance of so-called indirect genetic, or social effects, that is, the heritable effect of an individual on the phenotype of a conspecific. Not only in plants, but also in animals, indirect genetic effects are an important determinant of fitness, or, in agricultural systems, of yield. For example, a regular practice in dairy farming is the removal of the horns of cattle, to prevent them from harmful fighting. Horns are an individually selected competitive trait that decreases group productivity, and hornless breeds have been selected to increase group productivity. Likewise, in poultry, the beaks of chicken are trimmed. Recent work by Piter Bijma and others shows that selection for sociality can work. Whereas Denison discusses the classical work by Bill Muir on group selection on chicken, unfortunately, he does not discuss this recent work.

Not only can we apply evolutionary insights to agriculture, we can also obtain new insights from the study of nonhuman examples of agriculture. For example, leaf-cutting ants and macrotermitine termites cultivate fungi for food, and mycorrhizal plants and legumes cultivate microorganisms for minerals. Similar to human agriculture, the ‘farming’ host in these mutualisms faces the challenge of maximizing the yield of a group of symbionts, in the face of individual selection potentially disrupting group productivity. Not only may these systems provide useful ideas for human agriculture, performance of the symbionts of rhizobial and mycorrhizal plants is also of direct relevance for our own agriculture, as our most important crops belong to these groups.

In the final chapter, Denison argues that we should hedge our bets with a greater variety of crops and ideas. “Darwinian selection among ideas” requires variation among ideas and strong selection, and especially the right criteria, aiming at long-term goals. Governments can provide incentives by subsidies or taxes, and the criteria should be based on outcome, rather than on practice, because existing ideas on sustainable practices may not be the best. Denison testifies a strong belief in the market, be it regulated by governments that formulate the right outcomes. Farmers and researchers then create the best solutions, better than governments will ever be able to.

This book is a rich source of information for evolutionary biologists, biotechnologists, and agriculturalists. It illustrates important evolutionary principles in an accessible way, using the farm of brother Tom as a recurrent tangible example. Evolutionary concepts, such as kin selection and relatedness are explained clearly, and illustrated with many examples that can be used for teaching. I can recommend this book to all students of evolutionary biology and ecology who are not afraid of applications. In fact, I may want to recommend it even more strongly to all those researchers, institutes, and companies whose research aim it is to face the challenge of a growing world population that needs to be fed on a planet on which the climate is rapidly changing. Denison's arguments are convincing and we as humans may be missing out on a bright future if we ignore this book.

Last but not least: Denison breaks a lance for funding a high diversity of research, performed by many small research groups, rather than a few large. Like tall plants, big research groups shadow small research groups, and like nitrogen fertilization becomes less efficient at high doses, funding tends to become progressively less efficient. Let us hope funding agencies also read Denison's book.

Associate Editor: M. Wade