Astronomical observations reveal a major deficiency in our understanding of physics—the detectable mass is insufficient to explain the observed motions in a huge variety of systems given our current understanding of gravity, Einstein’s General theory of Relativity (GR). This missing gravity problem may indicate a breakdown of GR at low accelerations, as postulated by Milgromian dynamics (MOND). We review the MOND theory and its consequences, including in a cosmological context where we advocate a hybrid approach involving light sterile neutrinos to address MOND’s cluster-scale issues. We then test the novel predictions of MOND using evidence from galaxies, galaxy groups, galaxy clusters, and the large-scale structure of the universe. We also consider whether the standard cosmological paradigm ($\Lambda$CDM) can explain the observations and review several previously published highly significant falsifications of it. Our overall assessment considers both the extent to which the data agree with each theory and how much flexibility each has when accommodating the data, with the gold standard being a clear a priori prediction not informed by the data in question. Our conclusion is that MOND is favoured by a wealth of data across a huge range of astrophysical scales, ranging from the kpc scales of galactic bars to the Gpc scale of the local supervoid and the Hubble tension, which is alleviated in MOND through enhanced cosmic variance. We also consider several future tests, mostly at scales much smaller than galaxies.

The KBC void is a local underdensity with the observed relative density contrast δ ≡ 1 − ρ/ρ0 = 0.46 ± 0.06 between 40 and 300 Mpc around the Local Group. If mass is conserved in the Universe, such a void could explain the 5.3σ Hubble tension. However, the MXXL simulation shows that the KBC void causes 6.04σ tension with standard cosmology (ΛCDM). Combined with the Hubble tension, ΛCDM is ruled out at 7.09σ confidence. Consequently, the density and velocity distribution on Gpc scales suggest a long-range modification to gravity. In this context, we consider a cosmological MOND model supplemented with $11 \, \rm {eV}/c^{2}$ sterile neutrinos. We explain why this νHDM model has a nearly standard expansion history, primordial abundances of light elements, and cosmic microwave background (CMB) anisotropies. In MOND, structure growth is self-regulated by external fields from surrounding structures. We constrain our model parameters with the KBC void density profile, the local Hubble and deceleration parameters derived jointly from supernovae at redshifts 0.023−0.15, time delays in strong lensing systems, and the Local Group velocity relative to the CMB. Our best-fitting model simultaneously explains these observables at the $1.14{{\ \rm per\ cent}}$ confidence level (2.53σ tension) if the void is embedded in a time-independent external field of ${0.055 \, a_{_0}}$. Thus, we show for the first time that the KBC void can naturally resolve the Hubble tension in Milgromian dynamics. Given the many successful a priori MOND predictions on galaxy scales that are difficult to reconcile with ΛCDM, Milgromian dynamics supplemented by $11 \, \rm {eV}/c^{2}$ sterile neutrinos may provide a more holistic explanation for astronomical observations across all scales.

Many observed disc galaxies harbour a central bar. In the standard cosmological paradigm, galactic bars should be slowed down by dynamical friction from the dark matter halo. This friction depends on the galaxy’s physical properties in a complex way, making it impossible to formulate analytically. Fortunately, cosmological hydrodynamical simulations provide an excellent statistical population of galaxies, letting us quantify how simulated galactic bars evolve within dark matter haloes. We measure bar strengths, lengths, and pattern speeds in barred galaxies in state-of-the-art cosmological hydrodynamical simulations of the IllustrisTNG and EAGLE projects, using techniques similar to those used observationally. We then compare our results with the largest available observational sample at redshift z = 0. We show that the tension between these simulations and observations in the ratio of corotation radius to bar length is 12.62σ (TNG50), 13.56σ (TNG100), 2.94σ (EAGLE50), and 9.69σ (EAGLE100), revealing for the first time that the significant tension reported previously persists in the recently released TNG50. The lower statistical tension in EAGLE50 is actually caused by it only having five galaxies suitable for our analysis, but all four simulations give similar statistics for the bar pattern speed distribution. In addition, the fraction of disc galaxies with bars is similar between TNG50 and TNG100, though somewhat above EAGLE100. The simulated bar fraction and its trend with stellar mass both differ greatly from observations. These dramatic disagreements cast serious doubt on whether galaxies actually have massive cold dark matter haloes, with their associated dynamical friction acting on galactic bars.

El Gordo (ACT-CL J0102-4915) is an extremely massive galaxy cluster (M200 ≈ 3 × 1015 M⊙) at redshift z = 0.87 composed of two subclusters with a mass ratio of 3.6 merging at speed Vinfall ≈ 2500 km s−1. Such a fast collision between individually rare massive clusters is unexpected in Lambda cold dark matter (ΛCDM) cosmology at such high z. However, this is required for non-cosmological hydrodynamical simulations of the merger to match its observed properties. Here, we determine the probability of finding a similar object in a ΛCDM context using the Jubilee simulation box with a side length of $6 \, h^{-1}$ Gpc. We search for galaxy cluster pairs that have turned around from the cosmic expansion with properties similar to El Gordo in terms of total mass, mass ratio, redshift, and collision velocity relative to virial velocity. We fit the distribution of pair total mass quite accurately, with the fits used in two methods to infer the probability of observing El Gordo in the surveyed region. The more conservative (and detailed) method involves considering the expected distribution of pairwise mass and redshift for analogue pairs with similar dimensionless parameters to El Gordo in the past light-cone of a z = 0 observer. Detecting one pair with its mass and redshift rules out ΛCDM cosmology at 6.16σ. We also use the results of Kraljic and Sarkar to show that the Bullet Cluster is in 2.78σ tension once the sky coverage of its discovery survey is accounted for. Using a χ2 approach, the combined tension can be estimated as 6.43σ. Both collisions arise naturally in a Milgromian dynamics (MOND) cosmology with light sterile neutrinos.

We test Milgromian dynamics (MOND) using wide binary stars (WBs) with separations of 2–30 kAU. Locally, the WB orbital velocity in MOND should exceed the Newtonian prediction by |$\approx 20~{{\ \rm per\ cent}}$| at asymptotically large separations given the Galactic external field effect (EFE). We investigate this with a detailed statistical analysis of Gaia DR3 data on 8611 WBs within 250 pc of the Sun. Orbits are integrated in a rigorously calculated gravitational field that directly includes the EFE. We also allow line-of-sight contamination and undetected close binary companions to the stars in each WB. We interpolate between the Newtonian and Milgromian predictions using the parameter α_grav, with 0 indicating Newtonian gravity and 1 indicating MOND. Directly comparing the best Newtonian and Milgromian models reveals that Newtonian dynamics is preferred at 19σ confidence. Using a complementary Markov Chain Monte Carlo analysis, we find that |$\alpha _{\rm {grav}} = -0.021^{+0.065}_{-0.045}$|⁠, which is fully consistent with Newtonian gravity but excludes MOND at 16σ confidence. This is in line with the similar result of Pittordis and Sutherland using a somewhat different sample selection and less thoroughly explored population model. We show that although our best-fitting model does not fully reproduce the observations, an overwhelmingly strong preference for Newtonian gravity remains in a considerable range of variations to our analysis. Adapting the MOND interpolating function to explain this result would cause tension with rotation curve constraints. We discuss the broader implications of our results in light of other works, concluding that MOND must be substantially modified on small scales to account for local WBs.