colleagues’ analysis of the fossil specimen stands out is in their use of wide-ranging digi- tal reconstruction that corrects distortions of the fossil’s shape, and estimates missing parts. These digital methods are readily available and offer unique opportunities for research. However, many more shapes can be morphed and matched this way than would be possi- ble with conventional methods, and care is needed to generate only the most realistic options. It is therefore essential that any digi- tal reconstruction is carried out with detailed, first-hand knowledge of the original fossil, including how it is preserved and distorted. This point is particularly relevant with respect to the forward-projecting cheekbones of the newly discovered fossil. After reconstruction, this area looks smoothed, with hardly any sign of the original bone surface. One prominent aspect of MRD where the reconstruction could be improved is the front of the upper jaw. Here, digital processing resulted in a less accurate rep- resentation of what the characteristic, strongly projecting subnasal area would have looked like before the fossil was broken. MRD is a great addition to the fossil record of human evolution. Its discovery will substantially affect our thinking on the ori- gin of the genus Australopithecus specifically, and on the evolutionary family tree of early hominins more broadly. This work demon- strates the importance that a single fossil can have in palaeontology, something we should remember when we get puzzled looks and sighs from our colleagues in the experimen- tal biosciences regarding excitement about a sample size of n = 1. ■ Fred Spoor is at the Centre for Human Evolution Research, Natural History Museum, London SW7 5BD, UK, in the Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Leipzig, Germany, and in the Department of Anthropology, University College London, UK. e-mail: f.spoor@nhm.ac.uk 1. Wood, B. & Harrison, T. Nature 470, 347–352 (2011). 2. Kimbel, W. H. in Handbook of Paleoanthropology: Vol III: Phylogeny of Hominids (eds Henke, W. & Tattersall, I.) 2071–2105 (Springer, 2015). 3. Haile-Selassie, Y., Melillo, S. M., Vazzana, A., Benazzi, S. & Ryan T. M. Nature 573, 214–219 (2019). 4. Saylor, B. Z. et al. Nature 573, 220–224 (2019). 5. Kimbel, W. H., Johanson, D. C. & Rak, Y. Nature 368, 449–451 (1994). 6. Mongle, C. S., Strait, D. S. & Grine, F. E. J. Hum. Evol. 131, 28–39 (2019). 7. Zollikofer, C. P. et al. Nature 434, 755–759 (2005). 8. Walker, A., Leakey, R. E., Harris, J. M. & Brown, F. H. Nature 322, 517–522 (1986). 9. Kimbel, W. H. et al. J. Hum. Evol. 51, 134–152 (2006). 10. Spoor, F. Nature 521, 432–433 (2015). This article was published online on 28 August 2019. J A S O N T . B U R K E Atomic clocks are currently the gold standard of timekeeping. These devices measure time on the basis of transitions between two states of an atom. On pages 238 and 243, respectively, Masuda et al.1 and Seiferle et al.2 report progress towards a clock that instead uses transitions between two states of an atomic nucleus. Such a nuclear clock could outperform existing atomic timekeepers, and have applications in both fundamental and applied physics. Humans have been trying to measure the passage of time for thousands of years. From the sundial, to the hourglass, to the pocket watch, we have continually tried to improve our ability to quantify and standardize time. In the early 1900s, scientists struggled to define time consistently, and put forth vari- ous standards to help synchronize humanity. What was missing was a natural reference point that could be used, regardless of its loca- tion on Earth. We needed to define what a second truly meant: something fundamental that remains accurate and precise across all space, for all millennia. Scientists realized that the properties of atomic transitions are independent of loca- tion in space or time. This recognition led to the idea of using a known transition between two atomic states as a means to define time. If a standardized second could be defined as a specific and agreed-on number of atomic transitions, time could b e quantif ied. Researchers set out to do this in the 1930s, and by the end of the 1940s the world had its first atomic clock3. Over the past 70 years, atomic clocks have been continually improved and currently have a precision4 of about 1 part in 1018. But what if we could do better than these devices? What if we could make a clock that was 100,000 times smaller, was less susceptible to its environment and possibly had a precision of 1 part in 1019? An atomic nucleus, which is about 100,000 times smaller than an atom, could provide such a device5. Since 2003, researchers around the world have been trying to make a nuclear clock using the nucleus of a thorium-229 atom6. This nucleus, unlike all others that are known, has an excited state (called an isomeric state) that is only a few electronvolts (eV) in energy above its ground state7. As a result, the tran- sition between these two states is accessible using specialized lasers. The problem is that the exact energy of the isomeric state is cur- rently unknown. Masuda et al. and Seiferle et al. have made progress towards under- standing the exact character of the tho- rium-229 isomeric transition, by carrying out experiments that extend previous work7. In Masuda and colleagues’ experiment, a high-intensity X-ray beam was passed through a pair of silicon crystals that nar- rowed the energy range of the X-rays to 0.1 eV. These X-rays were then used to irra- diate a thorium-229 nucleus that was in the ground state (Fig. 1). The nucleus transi- tioned to a second excited state that has an energy much higher than that of the isomeric state. The narrow X-ray energy range allowed the authors to determine the exact energy of M E T R O L O G Y One tick closer to a nuclear clock Clocks that are based on the nucleus of a single thorium atom could be more precise than existing timekeepers. Such clocks have not yet been realized, but two experiments provide keys steps towards this goal. See Letters p.238 & p.243 Energy Second excited state Decay X-ray Electron Internal conversion X-ray excitation 29.19 keV 8.28 eV 0 eV Isomeric state Ground state Figure 1 | Low-energy states and transitions of the thorium-229 nucleus. Masuda et al.1 report a technique to produce nuclei of thorium-229 atoms in an excited state called an isomeric state. The authors irradiated a thorium-229 nucleus in the ground state with X-rays, which caused the nucleus to transition to a second excited state that has an energy of 29.19 keV (eV; electronvolts). The nucleus then decayed to the isomeric state. Seiferle et al.2 observed a process known as internal conversion, in which a thorium-229 nucleus in the isomeric state decayed to the ground state and the neutral atom emitted an electron. By studying the energy of emitted electrons, the authors estimated the energy of the isomeric state to be about 8.28 eV. These two studies could lead to ultraprecise clocks that are based on thorium-229 nuclei. 2 0 2 | N A T U R E | V O L 5 7 3 | 1 2 S E P T E M B E R 2 0 1 9 NEWS & VIEWSRESEARCH © 2019 Springer Nature Limited. All rights reserved. © 2019 Springer Nature Limited. All rights reserved. N O E M I D E R Z S Y How does the online hate ecosystem persist on social-media platforms, and what measures can be taken to effectively reduce its presence? On page 261, Johnson et al.1 address these questions in a captivating report on the behaviour of online hate communities that reside on multiple social-media platforms. The authors shed light on the structure and dynamics of online hate groups and, informed by the results, propose four policies to reduce hate content on online social media. We live in an age of high social inter- connectedness, whereby opinions shared in one geographical region do not remain spatially localized, but can spread rapidly around the globe thanks to online social media. The high speed of such diffusion poses problems for those policing hate speech, and creates opportunities for nefarious organiza- tions to share their messages and expand their recruiting efforts globally. When the policing of social media is inefficient, the online eco- system can become a powerful radicalizing instrument2. Understanding the mechanisms that govern hate-community dynamics is thus crucial to proposing effective measures to combat such organizations in this online battleground. Johnson et al. examined the dynamics of hate clusters on two social-media platforms, Facebook and VKontakte, over a period of a few months. Clusters were defined as online pages or groups that organized individuals who shared similar views, interests or declared purposes, into communities. These pages and groups on social-media platforms contain links to other clusters with similar content that users can join. Through these links, the authors established the network connections between clusters, and could track how members of one cluster also joined other clusters. Two clusters (groups or pages) were considered connected if they contained links to one another. The authors’ approach had the advantage of not requiring individual-level information about users who are members of clusters. Johnson et al. show that online hate groups are organized in highly resilient clusters. The users in these clusters are not geographically localized, but are globally interconnected by ‘highways’ that facilitate the spread of online hate across different countries, conti- nents and languages. When these clusters are attacked — for example, when hate groups are removed by social-media platform adminis- trators (Fig. 1) — the clusters rapidly rewire and repair themselves, and strong bonds are made between clusters, formed by users shared between them, analogous to covalent chemi- cal bonds. In some cases, two or more small clusters can even merge to form a large cluster, in a process the authors liken to the fusion of two atomic nuclei. Using their mathematical model, the authors demonstrated that banning hate content on a single platform aggravates online hate ecosystems and promotes the creation of clusters that are not detectable by platform policing (which the authors call ‘dark pools’), where hate content can thrive unchecked. O n l i n e s o c i a l - m e d i a pl at for ms are challenging to regulate, and policymakers have struggled to suggest practicable ways of reducing hate online. Efforts to ban and remove hate-related content have proved ineffective3,4. Over the past few years, the incidence of reports of hate speech online has been rising5, indicating that the battle against the diffusion of hateful content is being lost, an unsettling direction for the well-being and safety of our society. Furthermore, exposure to and engagement with online hate on social media has been suggested to promote offline aggression6, with some perpetrators of violent hate crimes reported to have engaged with such content7. Previous studies (for example, ref. 8) have considered hate groups as individual net- works, or considered the interconnected clusters together as one global network. In their fresh approach, Johnson and colleagues studied the interconnected structure of a community of hate clusters as a ‘network of networks’9–11, in which clusters are networks that are inter connected by highways. More- over, they propose four policies for effective intervention that are informed by the mecha- nisms their study revealed govern the structure and dynamics of the online-hate ecosystem. Currently, social-media companies must decide which content to ban, but often have to contend with overwhelming volumes of content and various legal and regulatory S O C I A L S C I E N C E The dynamics of online hate An analysis of the dynamics of online hate groups on social-media platforms reveals why current methods to ban hate content are ineffective, and provides the basis for four potential strategies to combat online hate. See Letter p.261 this second excited state: 29.19 keV. Finally, the nucleus decayed directly to the isomeric state. The approach of Masuda et al. could enable this state to be produced more effi- ciently than was previously possible. In Seiferle and colleagues’ experiment, a beam of thorium-229 ions was generated from the natural decay of uranium-233 ions. About 2% of the thorium ions were in the iso- meric state. These ions were then neutralized to allow them to decay to the ground state through a process called internal conversion. In this process, a nuclear decay that would typically produce a γ-ray instead causes the neutral atom to emit an electron (Fig. 1). However, internal conversion is complicated, because the electron can originate from many different energy levels in the neutral atom. To observe the ejected electrons from internal conversion, Seiferle and co-workers used a magnetic field to bend the trajec- tory of these particles towards an electron detector. They applied an electric field to the electrons until the voltage associated with this field was large enough to stop the electrons. The final voltage was equal to the initial energy of the electrons. Seiferle et al. then used a theoretical model to interpret the electron energy spectrum, which is the first energy spectrum observed from the decay products of the isomeric state. Their analysis indicated that the energy of the isomeric state is 8.28 ± 0.17 eV. Although the ultimate and groundbreaking goal of directly observing the thorium-229 isomeric transition remains elusive, sub- stantial progress continues to be made. The results of Masuda et al. and Seiferle et al. are key steps forward. Hopefully, the observation is not too far off, as teams of scientists race to make the world’s first nuclear clock, which would offer unprecedented precision. This finding would enable a whole host of experi- ments and discoveries in the decades to fol- low. For instance, a nuclear clock could have applications in dark-matter research8 and in the observation of possible variations in the fundamental constants of physics9. ■ Jason T. Burke is in the Nuclear and Particle Physics Group, Nuclear and Chemical Sciences Division, Physical and Life Sciences Directorate, Lawrence Livermore National Laboratory, Livermore, California 94550, USA. e-mail: burke26@llnl.gov 1. Masuda, T. et al. Nature 573, 238–242 (2019). 2. Seiferle, B. et al. Nature 573, 243–246 (2019). 3. Lyons, H. Instruments 22, 133–135 (1949). 4. Brewer, S. M. et al. Phys. Rev. Lett. 123, 033201 (2019). 5. Campbell, C. J. et al. Phys. Rev. Lett. 108, 120802 (2012). 6. Peik, E. & Tamm, C. Europhys. Lett. 61, 181–186 (2003). 7. Beck, B. R. et al. Phys. Rev. Lett. 98, 142501 (2007). 8. Derevianko, A. & Pospelov, M. Nature Phys. 10, 933–936 (2014). 9. Flambaum, V. V. Phys. Rev. Lett. 97, 092502 (2006). 1 2 S E P T E M B E R 2 0 1 9 | V O L 5 7 3 | N A T U R E | 2 0 3 NEWS & VIEWS RESEARCH © 2019 Springer Nature Limited. All rights reserved. © 2019 Springer Nature Limited. All rights reserved.