Two giant galaxy clusters, Abell 3391 and Abell 3395, are slowly colliding in the southern sky. Between them stretches a bridge of scorching gas, like a mist over a river, linking two banks. Astronomers have long wanted to know whether this "intergalactic bridge" hosts a magnetic field and how it behaves.
The field cannot be seen directly, but its trace can be caught. When radio waves from distant galaxies pass through magnetized gas, their plane of oscillation rotates slightly — an effect known as Faraday rotation. The stronger and more tangled the field, the greater the rotation. The astronomers aimed the MeerKAT radio telescope in South Africa at this region, collected signals from 434 distant galaxies, and built a magnetic field map from their rotations.
This contrast hints that the field in the bridge is not just present: it is aligned by some process. Perhaps gas flows streaming between the clusters stretch and straighten the magnetic field lines, much as a river aligns seaweed with the current. In the clusters themselves, the field is constantly shaken: hot gas churns, supernovae explode, and black holes at the centers of galaxies are active.
Such a "magnetic bridge" between clusters is a rare stroke of luck for astronomers. It offers a glimpse of how magnetic fields behave on the largest scales — where individual galaxies no longer matter, and only the cosmic web counts. Understanding these fields will help explain how giant structures grow and why charged particles in space move the way they do.
as written by the authors arXiv:2604.18735
Magnetic fields in cluster outskirts and the intercluster medium are poorly constrained because diffuse synchrotron emission is hard to detect at low surface brightness. Faraday rotation measures (RMs) of polarized background sources can probe foreground large-scale structure. The nearby interacting Abell 3391-3395 system hosts a well-established X-ray bridge, making it an excellent target for studying magnetization in the intercluster environment. We characterize the magnetized environment of Abell 3391/95 and its surroundings by constructing a dense RM grid from wide-field polarimetry. We observed Abell 3391/95 with MeerKAT in full polarization using a three-pointing mosaic. The data were calibrated with direction-independent and direction-dependent techniques and imaged using visibility-plane mosaicing for a large field of view at high sensitivity. Using Faraday synthesis, we formed Faraday cubes and measured RMs for polarized background sources. We defined on- and off-target regions using contours from a wavelet-filtered eROSITA image. We identified 434 polarized sources within the field, with a polarized source density ranging from about 30 sources per square degree in the outer regions to about 110 sources per square degree in the central field, and a field-averaged density of 73 sources per square degree. The clusters show a statistically significant enhancement of RM scatter relative to the off-target region. In contrast, the bridge shows comparatively low RM scatter, while an RM structure-function analysis on matched angular scales yields a tentative indication of larger RM differences in the bridge than off-target. Combined with low per-source depolarization, this suggests a bridge magnetic field relatively ordered on ~10 kpc scales, but less ordered on larger scales. The non-detection of diffuse synchrotron emission in the bridge yields improved upper limits on the emissivity.
The density of "background" galaxies used as probes reached 110 per square degree at the center of the field of view, as if a whole crowd of distant radio sources were packed into an area the size of a fingernail at arm's length.