MICROBIOME. A unified initiative to harness Earth's microbiomes.

507 SCIENCE sciencemag.org nels, generating a " bowtie " image. These very different patterns of deep critical zone structure are not easily explained with climate. The authors used a numerical model of the state of stress in an elastic rock mass into which a landscape has been carved (9) to calculate the pattern of expected cracking of the rock. The topographic stresses arise from both the topography itself, and the far-field horizontal stresses imposed by the tectonic setting (arrows in figure) constrained by an existing world map of stresses. As the far-field stresses are increased , the pattern of expected cracking morphs from the surface-parallel to bowtie patterns, capturing both end-members of the observed seismic images. This is indeed an encouraging result. Are we to believe their results? In many mountain ranges, the rock arriving in the near-surface zone is already riddled with flaws that have accumulated as it moved through the tectonic stress fields of present and past orogenies (10). To what degree does the presence of such preexisting flaws violate the assumption that the rock behaves as a uniform elastic medium? How well does the present state of stress reflect the long-term history of stress to which a rock has been subjected? One can also imagine situations in which other processes that generate near-surface cracks [for example, frost-cracking (11)], or that chemically weather the rock as it nears the surface (12), are instead the rate-limiting steps in damaging the rock. Whatever the answers, the results reported by St. Clair et al. will challenge the broader community to entertain a role for the state of stress imposed by the topography itself and its tectonic setting. They have also demonstrated the utility of classical geophysical methods and of a network of sites to test their ideas. ■ MICROBIOME A unified initiative to harness Earth's microbiomes Transition from description to causality and engineering D espite their centrality to life on Earth, we know little about how microbes (1) interact with each other, their hosts, or their environment. Although DNA se-quencing technologies have enabled a new view of the ubiquity and diversity of microorganisms, this has mainly yielded snapshots that shed limited light on microbial functions or community dynamics. Given that nearly every habitat and organism hosts a diverse constellation of micro-organisms—its " microbiome " — such knowledge could transform our understanding of the world and launch innovations in agriculture, energy, health, …