Nature | News

Dark matter’s tendrils revealed

Direct measurement of a dark-matter ‘filament’ confirms its existence in a galaxy supercluster.

Article tools

Rights & Permissions

Jörg Dietrich, University of Michigan/University Observatory Munich

Dark-matter filaments, such as the one bridging the galaxy clusters Abell 222 and Abell 223, are predicted to contain more than half of all matter in the Universe.

A ‘finger’ of the Universe’s dark-matter skeleton, which ultimately dictates where galaxies form, has been observed for the first time. Researchers have directly detected a slim bridge of dark matter joining two clusters of galaxies, using a technique that could eventually help astrophysicists to understand the structure of the Universe and identify what makes up the mysterious invisible substance known as dark matter.

According to the standard model of cosmology, visible stars and galaxies trace a pattern across the sky known as the cosmic web, which was originally etched out by dark matter — the substance thought to account for almost 80% of the Universe’s matter. Soon after the Big Bang, regions that were slightly denser than others pulled in dark matter, which clumped together and eventually collapsed into flat ‘pancakes’. “Where these pancakes intersect, you get long strands of dark matter, or filaments,” explains Jörg Dietrich, a cosmologist at the University Observatory Munich in Germany. Clusters of galaxies then formed at the nodes of the cosmic web, where these filaments crossed.

The presence of dark matter is usually inferred by the way its strong gravity bends light travelling from distant galaxies that lie behind it — distorting their apparent shapes as seen by telescopes on Earth. But it is difficult to observe this 'gravitational lensing' by dark matter in filaments because they contain relatively little mass.

Dietrich and his colleagues got around this problem by studying a particularly massive filament, 18 megaparsecs long, that bridges the galaxy clusters Abell 222 and Abell 223. Luckily, this dark bridge is oriented so that most of its mass lies along the line of sight to Earth, enhancing the lensing effect, explains Dietrich. The team examined the distortion of more than 40,000 background galaxies, and calculated that the mass in the filament is between 6.5 × 1013 and 9.8 × 1013 times the mass of the Sun. Their results are reported in Nature today1.

Mass equation

By examining X-rays from plasma in the filament, observed by the XMM-Newton spacecraft2, the team calculated that no more than 9% of the filament's mass could be made up of hot gas. The team's computer simulations suggest that roughly another 10% of the mass could be due to visible stars and galaxies. The bulk, therefore, must be dark matter, says Dietrich.

Mark Bautz, an astrophysicist at the Massachusetts Institute of Technology in Cambridge, notes that astrophysicists do not know precisely how visible matter follows the paths laid out by dark matter. “What’s exciting is that in this unusual system we can map both dark matter and visible matter together and try to figure out how they connect and evolve along the filament,” he says. Japan’s Astro-H X-ray space telescope, due to launch in 2014, will be able to characterize the ionization state and temperature of the plasma in the filament, which will help to discriminate between different models of how the structure formed.

Refining the technique could also help to pin down the identity of dark matter — whether it is a cold (slow-moving) particle or a warm (fast-moving) one, like a neutrino — because different particles will clump differently along the filament. The Euclid space mission, due to launch in 2019, will provide more lensing data. “This will complement direct dark-matter searches, for example at the Large Hadron Collider,” says Alexandre Refregier, a cosmologist at ETH Zurich, the Swiss Federal Institute of Technology in Zurich.

Journal name:
Nature
DOI:
doi:10.1038/nature.2012.10951

References

  1. Dietrich, J. P. et al. Nature http://dx.doi.org/10.1038/nature11224 (2012).

  2. Werner, N. et al. Astron. Astrophys. 482, L29L33 (2008).

For the best commenting experience, please login or register as a user and agree to our Community Guidelines. You will be re-directed back to this page where you will see comments updating in real-time and have the ability to recommend comments to other users.

Comments for this thread are now closed.

Comments

24 comments Subscribe to comments

  1. Avatar for More Info
    More Info

    The integral mass is increasing at least as fast as R. The mass is not converging to a limiting mass at the edge of the optical image

    addiction recovery

  2. Avatar for Rendall Portilla
    Rendall Portilla

    I do agree with Jon by the way, Rent a Car Romania

  3. Avatar for Rendall Portilla
    Rendall Portilla

    I do think that the establishes that dark matter does not lies in our 4D space-time but it lies in some 5th spatial dimension from where it casts its influence over ordinary matter. Gravity is the common thread between 5th spatial dimension and our normal 4D space Time. "Rent a Car Romaniaâ&#x80&#x9d:http://www.car-rent-romania.com

  4. Avatar for Jon Sehgal
    Jon Sehgal

    I know what you mean, but maybe you should look for multiple free nucleotide polymorphisms . I am sure that you will find new theories about that .

  5. Avatar for Abdullrahman Alabdully
    Abdullrahman Alabdully

    It would be useful and timely for Nature to do a feature on BSL3 and BSL4 lab protocols for assessing the psychological stability, over time, <a href="http://www.n99m.com/" title="Arab">Arab</a> of people who work in such labs.

  6. Avatar for Jacqueline Pierr Boaline
    Jacqueline Pierr Boaline

    I really that this form for the rotation curves implies that the mass is not centrally condensed, but that significant mass is located at large R. best uk directory integral mass is increasing at least as fast as R. The mass is not converging to a limiting mass at the edge of the optical image.

  7. Avatar for James T. Dwyer
    James T. Dwyer
    - Nicolaas Vroom,

    That is the section I referred to in my earlier comment &#8211 sorry my link did't paste so well. The complete quotation of conclusion 1 on page 485 is:
    "VIII. DISCUSSION AND CONCLUSIONS":
    "1. Most galaxies exhibit rising rotational velocities at the last measured velocity; only for the very largest galaxies are the rotation curves flat. Thus the smallest Sc's (i.e., lowest luminosity) exhibit the same lack of a Keplerian velocity decrease at large R as do the high-luminosity spirals. This form for the rotation curves implies that the mass is not centrally condensed, but that significant mass is located at large R. The integral mass is increasing at least as fast as R. The mass is not converging to a limiting mass at the edge of the optical image. The conclusion is inescapable that non-luminous matter exists beyond the optical galaxy."

    As indicated by this statement and evidenced throughout this and preceding research reports, it seems that their dark matter conclusion was not based on simulation studies but a simple comparison of characteristic "Keplerian" and actual galactic rotation curves. No charts of simulated or predicted rotation curves for specific galaxies studied are included.. The referenced Rubin et al. research report includes the actual rotation curves for each of the 21 Sc galaxies studied on page 477.

    As far as I can tell, the chart shown in the wikipedia entry is nothing more than an illustration of the basis for the perceived galaxy rotation problem.

    While the presence of dark matter within the galactic disc cannot be precluded, it would not meet the intended requirements to produce the observed relatively flat rotation curves of galactic disks. It is currently estimated that the amount of mass provided by dark matter configured within the peripheral dark matter halo (located between 100,000 and 300,000 light years from the galactic center) is about 10 times the mass provided by ordinary galactic matter. Please see Battaglia et al. (2005), The radial velocity dispersion profile of the Galactic halo: constraining the density profile of the dark halo of the Milky Way

    That research identifies more than 200 objects comprised of ordinary matter that orbit the Milky Way within its distant visible matter halo. These observed halo objects include satellite galaxies, globular clusters and old stars &#8211 discrete objects that crucially do comply with characteristic Keplerian diminishing rotational curves. I suggest that these objects in effect each independently orbit the collective mass of the galactic disc and bulge similarly to the sparse planets orbiting the Sun, which contains 99.86% of total Solar system mass.

    Thanks very much for the tip on underscores in link addresses.

  8. Avatar for Nicolaas Vroom
    Nicolaas Vroom

    The article by Rubin is at: http://adsabs.harvard.edu/abs/1980ApJ...238..471R
    I do not see to which page you refer. At page 485 of that article you can read: "The conclusion is inescapable that nonluminous matter exists beyond the optical galaxy." That means if you do a simulation without this extra nonluminous matter you are wrong. The isue is what is this nonluminous matter. The most obvious candidate is cold baryonic matter.
    The galaxy rotation curve url you mention is at: http://en.wikipedia.org/wiki/Galaxy_rotation_curve. For the underscore in the url you need "%5F" The problem is that I have a lot of objections against this document. For an overview go to: http://users.telenet.be/nicvroom/wik_galaxy_rotation_curve.htm

  9. Avatar for James T. Dwyer
    James T. Dwyer
    Nicolaas Vroom &#8211 you stated:
    "<em>You can use Newton's Law to study a Galaxy and our solar system. When you do the last you will see that the planets behave as Keppler's third law, but you can not use Keppler's third law to study a Galaxy because Keppler's third law starts from a point mass.</em>"

    I agree, but, please see my 2012-07-08 06:01 AM comment's quote of Rubin et al. (and search that seminal paper's references) - it was, unfortunately, concluded that undetected peripheral mass was necessary to produce the observed flat rotation curves - simply because the observed rotational velocities of galaxy disc objects remained flat at increasing radii rather than diminishing as predicted by Kepler's third law of planetary motion &#8211 actually only determined by empirical comparison with characteristic "Keplerian" rotation curves of planetary systems.

    Also please search http://en.wikipedia.org/wiki/ for the entry "Galaxy rotation curve" (this comment entry app can't handle underscores within URL link addresses, since it indicates formatted text)...

  10. Avatar for Nicolaas Vroom
    Nicolaas Vroom

    James Dwyer said: "NV - Your distinction between mass inside and outside the galaxy is moot, since only additional mass outside the visible galaxy would produce the observed galactic flat rotation curves (grv)" If you want to calculate a grv you have to use Newton's Law and take into account all 3D matter inside the bulge and the disk. That means all visible and invisible matter (including all invisible matter outside the visible disc) i.e. all hot and cold baryonic normal matter (i.e. all BH's stars planets dust and gass clouds). Only when this calculated grv does not match with the observed grv you can use nonbaryonic matter.
    You can use Newton's Law to study a Galaxy and our solar system. When you do the last you will see that the planets behave as Keppler's third law, but you can not use Keppler's third law to study a Galaxy because Keppler's third law starts from a point mass.

  11. Avatar for Vinod Sehgal
    Vinod Sehgal
    The article at one stage states that" after the Big Bang, regions that were slightly denser than others pulled in dark matter, which clumped together and eventually collapsed into flat ââ&#x82¬Ë&#x9cpancakesââ&#x82¬â&#x84¢. ââ&#x82¬Å&#x93Where these pancakes intersect, you get long strands of dark matter, or filaments,ââ&#x82¬Â&#x9d explains JÃ&#x83¶rg Dietrich, a cosmologist at the University Observatory Munich in Germany."

    At another stage, article also states" Mark Bautz, an astrophysicist at the Massachusetts Institute of Technology in Cambridge, notes that astrophysicists do not know precisely how visible matter follows the paths laid out by dark matter."

    From the Ist statement, it appears that after BB, slightly denser regions of ordinary matter attracted dark matter towards it implying it is the dark matter which follows ordinary matter. While second statement clearly reveals that ordinary matter follows dark matter though astronomers do not know precisely how ordinary matter follows dark matter.

    So which follows what ? Ordinary matter follows dark matter or dark matter follows ordinary matte?. Astronomers estimate mass of dark matter to be 80% of ordinary matter, therefore, their gravitation effects should also be in that proportion. As such, an any area of universe, it the ordinary which should follow the pattern of dark matter during all the cosmological times.

    In an any area of universe, distribution o of ordinary matter mass can not lead to inference that dark matter shall also be distributed in that pattern ( of ordinary pattern). Gravitational effects in any area, as inferred from motion of ordinary matter, and gravitational lensing effect in any area of universe may be function of 3 factors1) visible baryonic matter 2) Invisible baryonic matter &#8211 MACHOS 3) Dark matter. Since knowledge of astronomer is limited up to1) only, therefore, he can not assign all balance mass to 3) only. There may be some contribution from 2) also

    Since quite long, scientists have been sensing the presence of dark matter on visible ordinary matter in 4 D space-Time, in variety of ways but not able to fix dark matter in either WIMPs or thru any other agent of ordinary matter. MACHOs and other objects are not as numerous to provide for the 80% of invisible mass. Otherwise also, there is no mechanism to suggest that 80% of matter since BB might have been converted into BHs and other invisible MACHO objects.. This establishes that dark matter does not lies in our 4D space-time but it lies in some 5th spatial dimension from where it casts its influence over ordinary matter. Gravity is the common thread between 5th spatial dimension and our normal 4D space Time. On detection of gravitons or gravity waves, scientists may enter the 5th dimension and lay hand over dark matter. Till then, despite spending billions on WIMP detectors, they will move in wilderness in 4D space-Time in search of all elusive dark matter

  12. Avatar for Vinod Sehgal
    Vinod Sehgal
    The article at one stage states that" after the Big Bang, regions that were slightly denser than others pulled in dark matter, which clumped together and eventually collapsed into flat â&#x80&#x98pancakesâ&#x80&#x99. â&#x80&#x9cWhere these pancakes intersect, you get long strands of dark matter, or filaments,â&#x80&#x9d explains Jörg Dietrich, a cosmologist at the University Observatory Munich in Germany."

    At another stage, article also states" Mark Bautz, an astrophysicist at the Massachusetts Institute of Technology in Cambridge, notes that astrophysicists do not know precisely how visible matter follows the paths laid out by dark matter."

    From the Ist statement, it appears that after BB, slightly denser regions of ordinary matter attracted dark matter towards it implying it is the dark matter which follows ordinary matter. While second statement clearly reveals that ordinary matter follows dark matter though astronomers do not know precisely how ordinary matter follows dark matter.

    So which follows what ? Ordinary matter follows dark matter or dark matter follows ordinary matte?. Astronomers estimate mass of dark matter to be 80% of ordinary matter, therefore, their gravitation effects should also be in that proportion. As such, an any area of universe, it the ordinary which should follow the pattern of dark matter during all the cosmological times.

    In an any area of universe, distribution o of ordinary matter mass can not lead to inference that dark matter shall also be distributed in that pattern ( of ordinary pattern). Gravitational effects in any area, as inferred from motion of ordinary matter, and gravitational lensing effect in any area of universe may be function of 3 factors1) visible baryonic matter 2) Invisible baryonic matter &#8211 MACHOS 3) Dark matter. Since knowledge of astronomer is limited up to1) only, therefore, he can not assign all balance mass to 3) only. There may be some contribution from 2) also

    Since quite long, scientists have been sensing the presence of dark matter on visible ordinary matter in 4 D space-Time, in variety of ways but not able to fix dark matter in either WIMPs or thru any other agent of ordinary matter. MACHOs and other objects are not as numerous to provide for the 80% of invisible mass. Otherwise also, there is no mechanism to suggest that 80% of matter since BB might have been converted into BHs and other invisible MACHO objects.. This establishes that dark matter does not lies in our 4D space-time but it lies in some 5th spatial dimension from where it casts its influence over ordinary matter. Gravity is the common thread between 5th spatial dimension and our normal 4D space Time. On detection of gravitons or gravity waves, scientists may enter the 5th dimension and lay hand over dark matter. Till then, despite spending billions on WIMP detectors, they will move in wilderness in 4D space-Time in search of all elusive dark matter

  13. Avatar for James T. Dwyer
    James T. Dwyer
    - correction &#8211 the preceding comment should have stated:
    The fundamental error is the presumption that galaxies must comply with the laws of planetary motion and therefore *should exhibit <em>diminishing rotation curves as radial distance increases*</em> that is the root cause of the perceived observational discrepancy with any physical laws.

    I apologize for any confusion this might have caused.

  14. Avatar for James T. Dwyer
    James T. Dwyer
    Nicolaas Vroom &#8211 Your distinction between mass inside and outside the galaxy is moot, since only additional mass <em>outside</em> the visible galaxy would produce the observed galactic flat rotation curves if it is erroneously assumed that the laws of planetary motion and characteristic Keplerian rotation curves should apply to spiral galaxies. If one hopes that the perceived 'missing' mass is contained in an enormous cloud enveloping the galaxy composed of either WIMPs or MACHOs, any increase in mass within the visible galaxy would have to be offset by even more compensatory mass outside the visible galaxy. On that issue Rubin et al. were correct.

    The fundamental error is the presumption that galaxies must comply with the laws of planetary motion and therefore should exhibit flat rotation curves that is the root cause of the perceived observational discrepancy with any physical laws.

    Proper evaluation of the rotational characteristics of spiral galaxies should not presume or expect that an enormous, gravitationally bound, highly distributed object of mass >100,000 light years in diameter, composed of hundreds of billions of stars and many other discrete objects of mass must be identical to the characteristics empirically derived through observation of the Solar system, in which 99.86% of total system mass is contained within the Sun.

    While I can't assess the extremely complex evaluation methods used to infer the presence of dark matter at larger scales, the evaluations of galactic rotation that established the inferred existence of any dark matter were faulty.

  15. Avatar for Nicolaas Vroom
    Nicolaas Vroom

    Keplers third law cannot be used to study flat galaxy rotation curves. See For example: http://www.astro.cornell.edu/academics/courses/astro201/rotation_curves.htm
    I agree "partly" with your quote that <em>The conclusion is inescapable that non-luminous matter exists beyond the optical galaxy.</em> IMO this should have been: within and beyond.
    The true question to answer is: how much of this non-luminous matter is baryonic and how much non-baryonic. The two documents mentioned do not answer this question. IMO you can only assume non-baryonic if you are sure that you have accounted for all baryonic matter i.e. all planet sized objects and gas clouds.
    For more detail study: http://users.telenet.be/nicvroom/dark_mat.htm

  16. Avatar for James T. Dwyer
    James T. Dwyer
    Nicolaas Vroom &#8211 you stated: "The same questions exists if you want to simulate a galaxy rotation curve. Such a simulation is clearly wrong if only type 1 matter is included."

    That was the conclusion based on early studies that only identified that spiral galaxies did not comply with "Keplerian rotation curves" illustrating the rotational velocity|radial distance of planets in the Solar system. Please see the seminal research report &#8211 Rubin, et al., (1980), "Rotational properties of 21 SC galaxies with a large range of luminosities and radii, from NGC 4605 /R = 4kpc/ to UGC 2885 /R = 122 kpc/ ", http://dx.doi.org/10.1086/158003
    http://adsabs.harvard.edu/abs/1980ApJâ&#x80¦238..471R
    As stated in section VIII. "DISCUSSION AND CONCLUSIONS":
    "1. Most galaxies exhibit rising rotational velocities at the last measured velocity; only for the very largest galaxies are the rotation curves flat. Thus the smallest Scâ&#x80&#x99s (i.e., lowest luminosity) exhibit the same lack of a Keplerian velocity decrease at large R as do the high-luminosity spirals. This form for the rotation curves implies that the mass is not centrally condensed, but that significant mass is located at large R. The integral mass is increasing at least as fast as R. The mass is not converging to a limiting mass at the edge of the optical image. The conclusion is inescapable that non-luminous matter exists beyond the optical galaxy."

    For a more appropriate application of physics in the evaluation of spiral galaxy rotation, please see:
    Feng & Gallo, (2011), "Modeling the Newtonian dynamics for rotation curve analysis of thin-disk galaxies", doi:10.1088/1674-4527/11/12/005 http://arxiv.org/abs/1104.3236

  17. Avatar for Nicolaas Vroom
    Nicolaas Vroom

    In this whole discussion there are 3 types of matter involved:
    1. Visible (ordinary) baryonic matter. 2. Invisible baryonic matter and 3. Invisible non-baryonic matter. The types 2 and 3 are called darkmatter. The question is how much of each is involved in Abell 222. in Abell 223 and in the bridge linking the two. The issue is if there is specific reason to assume that type 3 is involved. The original article does not mention this.
    The same questions exists if you want to simulate a galaxy rotation curve. Such a simulation is clearly wrong if only type 1 matter is included. This is like simulating the solar system without including the Kuiper belt which belongs to type 2. Present observations reveal that there is much more type 2 matter in our Galaxy than original thought. This makes the necessity of type 3 matter and the prove of its existance less "solid".

  18. Avatar for Robert L. Oldershaw
    Robert L. Oldershaw

    The observational evidence for dark matter is overwhelming.

    The observational evidence for "WIMPs" = 0.

    These two facts, supported by nearly all professional physicists and astrophysicists, have been true for decades.

    Robert L. Oldershaw
    http://www3.amerst.edu/~rloldershaw
    Discrete Scale Relativity

  19. Avatar for James T. Dwyer
    James T. Dwyer
    brown paul &#8211 well put, but while other observational evidence, primarily extremely complex gravitational analyses of gravitational lensing effects, have been attributed to dark matter, I think its highly doubtful that any could have so successfully established the justification for dark matter's existence as well as the simple analyses of galaxy rotation. The attribution of gravitational lensing effects to dark matter was almost completely reliant on the 'galaxy rotation problem' having established a perceived requirement for the existence of dark matter, even though its has not been identified much less directly detected.

    To put very simply, it was erroneously expected that Kepler's laws of planetary motion must also apply to spiral galaxies, and that their rotation curves (rotational velocity|radial distance) must comply with those characteristic of the Solar system (unlike spiral galaxies, 99.86% of total Solar system mass is contained within the Sun). This premise was invalid.

    Moreover, gravitational theories do not require modification, only more appropriate application. Please see
    Feng & Gallo, (2011), "Modeling the Newtonian dynamics for rotation curve analysis of thin-disk galaxies", doi:10.1088/1674-4527/11/12/005 http://arxiv.org/abs/1104.3236

    Also see http://www.nature.com/news/galaxy-clusters-caught-in-motion-1.10920 for some previously unidentified factors that might be significant in the production of minute weak gravitational lensing effects by galaxy clusters.

  20. Avatar for Bob Guy
    Bob Guy

    maybe the two of you should get a room

  21. Avatar for Eric Gisse
    Eric Gisse

    Robert, why are you re-living an argument you have lost?

    I don't see why you are bringing up Hawkins' discredited works. Hawkins' theories on quasar variability have been rather clearly falsified in the literature (eg, astro-ph:0411348 ) and you have had this all explained to you before <sup>1</sup>. Note how you have no technical response to what is published in the literature or what arguments have been given to you in the newsgroups.

    Feel free to proclaim bias as much as you want. You have no technical response, as your methodology is "ignore the arguments, and repeat your assertions".

    Thanks for playing.

    <sup>1</sup> :

    http://groups.google.com/group/sci.astro.research/msg/f3e59845c58c2889?dmode=source
    http://groups.google.com/group/sci.astro.research/msg/280d203025cf7a66?dmode=source

  22. Avatar for Robert L. Oldershaw
    Robert L. Oldershaw

    PS: The recently launched NuSTAR X-ray telescope has a decent chance of detecting the vast population of isolated "primordial" black holes predicted by Discrete Scale Relativity for a galactic dark matter.

    In a new high energy range, NuSTAR has 10 times the resolution and 100 times the sensitivity of previous X-ray telescopes.

    NuSTAR is scheduled to begin sientific operations in about 10 days, and one of its primary missions is to quantify the abundance of stellar-mass holes in our Galaxy. If Discrete Scale Relativity's definitive prediction of trillions of isolated stellar-mass black holes is correct, then nuSTAR should at the very least see the high-mass tail of the population, which would still greatly exceed present conventional black hole abundance estimates.

    Such a detection would be something more impressive and tangible than a certain very indirectly inferred and somewhat anomalous resonance.

    Robert L. Oldershaw
    http://www3.amherst.edu/rloldershaw

  23. Avatar for James T. Dwyer
    James T. Dwyer

    The article states:
    "The team examined the distortion of more than 40,000 background galaxies, and calculated that the mass in the filament is between 6.5 Ã&#x97 1013 and 9.8 Ã&#x97 1013 times the mass of the Sun. Their results are reported in Nature today."

    What an exceedingly difficult task to estimate the total mass in the region from the exceedingly minute distortions identified by statistical analysis of the topography of 40,000 galaxies. Any estimation error in the evaluation of galaxy distortions would skew the estimated dark matter necessary to produce the estimated gravitational effects. I wonder if the results were confirmed using some independent method?

  24. Avatar for Robert L. Oldershaw
    Robert L. Oldershaw

    "Refining the technique could also help to pin down the identity of dark matter â&#x80&#x94 whether it is a cold (slow-moving) particle or a warm (fast-moving) one, like a neutrino â&#x80&#x94 because different particles will clump differently along the filament."

    Or: stellar-mass and planetary-mass black holes as definitively predicted by Discrete Scale Relativity.

    You forgot to mention that primordial black holes are still quite viable candidates for the galactic scale dark matter.

    Mythological "WIMP" particles, or any other type of subatomic particles, are NOT the "only game in town", and the only "WIMP miracle" is the fact that they still have such a following after 40 years of complete no-shows.

    Robert L. Oldershaw
    http://www3.amherst.edu/~rloldershaw
    Discrete Scale Relativity
    Fractal Cosmology

Taking a gamble

prediction-markets

The power of prediction markets

Scientists are beginning to understand why these ‘mini Wall Streets’ work so well at forecasting election results — and how they sometimes fail.

Newsletter

The best science news from Nature and beyond, direct to your inbox every day.

The polling crisis

election-polling

How to tell what people really think

This year’s US presidential election is the toughest test yet for political polls as experts struggle to keep up with changing demographics and technology.

Mitochondrial replacement

mitochondrial-replacement

Reports of 'three-parent babies' multiply

Claims of infants created using mitochondrial-replacement techniques stir scientific and ethical debate.

US presidential race

Trump-supporters

The scientists who support Donald Trump

Science policy fades into background for many who back Republican candidate in US presidential race.

ExoMars

lost-mars-lander

Europe’s probe feared lost on Mars

Sister craft successfully enters Martian orbit but loses contact with Schiaparelli lander.

Nature Podcast

new-pod-red

Listen

This week, making egg cells in a dish, super-bright flares in nearby galaxies, and trying to predict the election.

Science jobs from naturejobs