NEWS & VIEWS RESEARCH consistent with the emerging idea that belowground plant inputs to soils are the dominant precursors for the formation of soil organic matter4,7. Pinpointing which mechanism explains Averill and colleagues’ results will require more data and involve challenges common to all large observational data sets, includ­ ing unobserved variables and spurious cor­ relations. Perhaps different mycorrhizal associations reflect adaptations to environ­ mental conditions, as opposed to being the cause of ecosystem differences. For example, in colder climates, where the cold slows the decay of organic matter and trees produce tough leaves that are hard to break down, EM fungi might dominate simply because of their ability to acquire nitrogen from organic matter8,9. In Averill and colleagues’ analyses, the strength of the mycorrhizal effect depends on the amount of soil nitrogen. To investigate this dependency, I used their model results to calculate organic-matter stores in tem­ perate and tropical forests, where the myco­ rrhizal types co-occur and where the authors conclude that EM-dominated forests have 1.3 times more carbon per unit nitrogen. At the low end of the authors’ nitrogen-content range (0.2 kilograms of nitrogen per square metre), EM-dominated forests actually have less (0.96 times) carbon than AM forests, and at 1.0 kg N m–2, below which many of the observations fall, they have 1.21 times more. It is not until soil nitrogen reaches values at the upper end of their observations (3 kg N m–2) that carbon stores are 1.3 times greater in EMdominated forests, a pattern consistent with the idea that the strength of the mycorrhizal effect is strongly dependent on soil-nutrient availability6. Despite the need to further explore such nuances, Averill and colleagues’ findings have important implications for the way we manage land resources in the face of a changing carbon cycle and climate. We depend on model projections to inform strategies to preserve our natural resources, yet the relevant models have been developed on the basis of an understand­ ing of soil dynamics that is increasingly shown to be wanting10. Climate, soil texture and plant productivity drive soil organic-matter storage in these models11 but were found by Averill et al. not to play a determining part in organicmatter levels. Their finding that it is instead the relative dominance of trees associating with different mycorrhizal fungi that corre­ lates with the amount of soil organic matter highlights the need to consider how local-scale biotic interactions shape global and regionalscale carbon dynamics. ■ Mark A. Bradford is in the School of Forestry and Environmental Studies, Yale University, New Haven, Connecticut 06511, USA. e-mail: [email protected]

1. Statement on Signing the Soil Conservation and Domestic Allotment Act, 1 March 1936. American Presidency Project http://www.presidency.ucsb. edu/ws/?pid=15254. 2. Jobbágy, E. G. & Jackson, R. B. Ecol. Appl. 10, 423–436 (2000). 3. Denman, K. L. et al. in Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (eds Solomon, S. D. et al.) Ch. 7 (Cambridge Univ. Press, 2007). 4. Schmidt, M. W. I. et al. Nature 478, 49–56 (2011). 5. Averill, C., Turner, B. L. & Finzi, A. C. Nature 505, 543–545 (2014).

6. Orwin, K. H., Kirschbaum, M. U. F., St John, M. G. & Dickie, I. A. Ecol. Lett. 14, 493–502 (2011). 7. Clemmensen, K. E. et al. Science 339, 1615–1618 (2013). 8. Phillips, R. P., Brzostek, E. & Midgley, M. G. New Phytol. 199, 41–51 (2013). 9. Johnson, N. C., Angelard, C., Sanders, I. R. & Kiers, E. T. Ecol. Lett. 16, 140–153 (2013). 10. Wieder, W. R., Bonan, G. B. & Allison, S. D. Nature Clim. Change 3, 909–912 (2013). 11. Bonan, G. B., Hartman, M. D., Parton, W. J. & Wieder, W. R. Global Change Biol. 19, 957–974 (2013). This article was published online on 8 January 2014.

SO L A R SYST E M

Evaporating asteroid The asteroid Ceres has been thought to contain abundant water. Observations acquired with the Herschel Space Observatory now show that this Solar System object is spewing water vapour from its surface. See Letter p.525 HUMBERTO CAMPINS & CHRISTINE M. COMFORT

W

riting in this issue, Küppers et al.1 report that Ceres — a dwarf planet or the largest asteroid in the Solar System, depending on the definition used — is releasing water vapour from its surface at a rate of about 2 × 1026 molecules, or 6 kilograms, per second. The presence and abundance of water in asteroids2,3 are relevant to many areas of research on the Solar System, ranging from the origin of water and life on Earth to the largescale migration of giant planets such as Jupiter. Water has been suspected of being a sig­ nificant component of Ceres for more than 30 years4. But it is only now that observations obtained by Küppers et al., using the European

Space Agency’s Herschel Space Observatory, have allowed the direct identification of water molecules escaping from two regions on the surface of this object (Fig. 1). The authors’ result backs up previous indirect observational evidence5,6 for water in this planetary body, and is particularly timely given that NASA’s Dawn spacecraft7 will soon visit Ceres, fresh from its successful mission to another intrigu­ ing small world, the asteroid Vesta. One of the most puzzling questions about the origin and evolution of asteroids is why Vesta and Ceres are so different. They are both located in the main asteroid belt, between the orbits of Mars and Jupiter, and their orbits are quite close to each other: about 2.4 and 2.8 astronomical units from the Sun, respectively (1 astronomical unit is the mean

Figure 1 | Artist’s impression of the asteroid Ceres.  Küppers et al.1 have discovered water vapour emanating from two regions on the surface of the asteroid. (Figure adapted from an illustration by Chris Butler/SPL.) 2 3 JA N UA RY 2 0 1 4 | VO L 5 0 5 | N AT U R E | 4 8 7

© 2014 Macmillan Publishers Limited. All rights reserved

RESEARCH NEWS & VIEWS Sun–Earth distance). Yet these objects are opposites in terms of their composition and appearance. Whereas Vesta has experienced extensive heating and volcanic eruptions that covered the entire asteroid, Ceres’ surface and interior have not reached temperatures high enough to melt rocks. Interestingly, a greater abundance of water in Ceres than Vesta may have been a crucial factor in producing the two bodies’ radically different final states8. The source of the water vapour observed by Küppers and colleagues may be related to the process of heat dissipa­ tion that precluded the melting of rocks in Ceres. More specifically, one of the proposed production mechanisms for the water vapour being released from Ceres involves the melt­ ing of subsurface ice that then flows to the sur­ face and evaporates into space. Water vapour has a high capacity to transport heat, and so, during the formation of Ceres about 4.6 bil­ lion  years ago, the sublimation of water ice might have efficiently dissipated the interior heat into space. This would have stopped Ceres from ending up with an igneous surface like that of Vesta. If this is indeed what happened during the formation of the two asteroids, one may ask: why did Ceres form with (and why does it still contain) more water than Vesta? It is most likely that Ceres formed in a colder outer region of the nascent Solar System than Vesta, beyond the snow line — the distance from the young Sun at which temperatures were low enough for water to form ice. But this hypoth­ esis raises the question of why Ceres and Vesta are so close to each other now. It has been sug­ gested that, soon after the formation of the asteroids and the planets, mixing of material from the inner and outer regions of the Solar System occurred. Such mixing would have been caused by migration of the orbits of Jupi­ ter and the other giant planets9, and that could have moved Ceres and Vesta from distant for­ mation sites to their current locations. One of the first clues that giant planets in the Solar System could undergo significant migration came from the discovery10 in 1995 that certain giant exoplanets are closer to their hosts than Mercury is to the Sun — orbiting at distances at which they could not have formed. The best explanation for these ‘hot Jupiters’ is that they formed far from their host star and that later their orbits reduced dramatically. Planetary migration has since been used to explain several puzzling observations. For example, the migration of Jupiter may have been responsible for the different composi­ tional groups observed within the asteroid belt9 and for a period of extensive impacts — known as the Late Heavy Bombardment — that occurred about 4 billion years ago11–13. According to this scenario, as the giant planets migrated, they disturbed populations of small rocky and icy bodies (asteroids and comets), which hit the early Earth and Moon. These

small bodies delivered organic molecules and water to Earth. Hence, early impacts by aster­ oids and comets might have played a consider­ able part in the origin and evolution of life on our planet. Küppers and colleagues’ detection of water vapour around Ceres and, more generally, our knowledge of Ceres and Vesta, are consist­ ent with emerging views of how giant-planet migration and other related processes shaped the Solar System’s early history. But the pieces of the puzzle of Solar System formation do not fit perfectly, and more is likely to be discov­ ered through further studies of the miniature worlds that we call asteroids. ■ Humberto Campins and Christine M. Comfort are in the Department of Physics and Astronomy, University of Central Florida, Orlando, Florida 32816-2385, USA. e-mail: [email protected]

1. Küppers, M. et al. Nature 505, 525–527 (2014). 2. Campins, H. et al. Nature 464, 1320–1321 (2010). 3. Rivkin, A. S. & Emery, J. P. Nature 464, 1322–1323 (2010). 4. Lebofsky, L. A. Mon. Not. R. Astron. Soc. 182, 17–21 (1978). 5. A’Hearn, M. F. & Feldman, P. D. Icarus 98, 54–60 (1992). 6. Rivkin, A. S., Howell, E. S., Vilas, F. & Lebofsky, L. A. in Asteroids III (eds Bottke, W. F. Jr, Cellino, A., Paolicchi, P. & Binzel, R. P.) 235–253 (Univ. Arizona Press, 2002). 7. Russell, C. T. & Raymond, C. A. Space Sci. Rev. 163, 3–23 (2011). 8. McCord, T. B. & Sotin, C. J. Geophys. Res. 110, E05009 (2005). 9. Walsh, K., Morbidelli, A., Raymond, S., O’Brien, D. & Mandell, A. Meteorit. Planet. Sci. 47, 1941–1947 (2012). 10. Mayor, M. & Queloz, D. Nature 378, 355–359 (1995). 11. Tsiganis, K., Gomes, R., Morbidelli, A. & Levison, H. F. Nature 435, 459–461 (2005). 12. Morbidelli, A., Levison, H. F., Tsiganis, K. & Gomes, R. Nature 435, 462–465 (2005). 13. Gomes, R., Levison, H. F., Tsiganis, K. & Morbidelli, A. Nature 435, 466–469 (2005).

STEM C EL L S

Sex specificity in the blood Haematopoietic stem cells, from which blood cells originate, are shown to respond to oestrogen and divide more frequently in female mice than in males, probably preparing females for the increased demand for blood in pregnancy. See Letter p.555 DENA S. LEEMAN & ANNE BRUNET

M

ales and females exhibit differences not only in reproductive organs, but also in sexually dimorphic tissues such as the mammary gland, brain and mus­ cle. In such tissues, the activity of stem cells, which self-renew and produce differentiated cells for tissue maintenance and repair, differs between males and females1–4. A fundamental yet unexplored question is whether the stem cells of tissues without conspicuous sex dif­ ferences, such as the blood or gut, also exhibit sexually dimorphic function. On page 555 of this issue, Nakada et al.5 find that haematopoi­ etic stem cells (HSCs), which form the blood and immune system, do differ between male and female mice. The authors show that female HSCs respond to long-range oestrogen signals in a manner that seems to help mothers meet the haematopoietic demands of pregnancy. HSCs reside in the bone marrow and produce all blood cells, which in turn mediate processes ranging from immunity to clotting to oxygen transport. Nakada and colleagues find that, under basal conditions, female HSCs and their immediate progeny, multipotent progenitor cells (MPPs), divide more frequently than male HSCs, and generate more erythroid progenitors

4 8 8 | N AT U R E | VO L 5 0 5 | 2 3 JA N UA RY 2 0 1 4

© 2014 Macmillan Publishers Limited. All rights reserved

(the cells that give rise to red blood cells). Despite the increased frequency of divi­ sion in female HSCs, males and females have the same basal number of HSCs and a similar cellular composition in the bone marrow and spleen (an organ colonized by haematopoietic cells). The authors suggest that female HSCs undergo more asymmetric divisions in which one daughter cell remains a stem cell and the other differentiates along the red blood cell lineage, and that these newly produced eryth­ roid progenitors undergo cell death at a higher frequency (Fig. 1). These differences may explain why the sexual dimorphism of HSCs has not previously been observed. During pregnancy, however, the authors observed a further increase in HSC prolif­ eration, an expansion of the number of HSCs in the bone marrow and spleen, and more erythroid cells in the spleen. Thus, it seems that female HSCs may be ‘primed’ for the increased demand for blood during pregnancy. Nakada et al. identify oestrogen as the causal agent for HSC sexual dimorphism (Fig. 1). They find that ovariectomy or pharmacologi­ cal inhibition of aromatase (an enzyme nec­ essary for oestrogen synthesis) reduced the percentage of proliferating HSCs and MPPs in females, whereas injection of oestradiol (the