If every galaxy became a single dot, would the Universe look like evenly scattered sugar? Maps of large volumes show something else: concentrations connect into long structures, with large, sparsely populated regions between them.
The cosmic web is the large-scale network of matter: nodes, filaments and sheets surround underdense voids. Galaxies mark part of this structure with their light; gas lies between them, and dark matter has an important gravitational role. The name describes structure, not a giant net stretched through space. ESA: matter distribution and cosmic structure.
One dot on this map may represent an entire galaxy
In a night-sky photograph, a dot often means a star. On a cosmic-web map, many points represent galaxies, each of which contains stars itself. We have changed scale, from the structure of one galaxy to the arrangement of many.
Galaxy clusters occupy the densest nodes. Filaments, or threads, stretch between them: elongated regions with elevated matter density. Broader flattened structures are called walls or sheets. Between these regions lie cosmic voids. They have lower density and few galaxies, but need not be absolutely empty. ESA: nodes of the web, Bond, Kofman and Pogosyan: a network of filaments and sheets.
Foam offers a useful everyday comparison: it has cells, walls and intersections. But space contains no soap film, surface tension or identical cells. Matter is distributed in three dimensions, and boundaries are defined using data and a chosen method.
Gravity amplified small initial differences
The early Universe was much more homogeneous than today’s, but not perfectly uniform. Slightly denser regions attracted surrounding matter. Over time the differences grew: dense regions became more prominent, while underdense ones lost matter relative to their surroundings. This is gravitational instability: a small initial deviation can grow. ESA: from a nearly smooth distribution to structure.
Matter did not fall equally from all directions. Initial distribution and unequal gravitational deformations favored sheets, filaments and nodes. Numerical models follow this growth from small fluctuations; researchers then compare the resulting network’s statistics with observations. It is not a story of someone tying ready-made galaxies together with ropes. Bond et al., 1996: the physical origin of the web.
In the standard cosmological model, dark matter supplies a substantial part of the gravitational framework. Ordinary gas also moves in that field, heats, cools and contributes to feeding galaxies. Dark matter is distinct from gas and dust: its distribution is inferred from gravitational effects, rather than photographed by its own light. ESA: the dark cosmic web.
A sky photograph alone cannot reveal the network
Two galaxies can appear side by side in an image yet be very far apart in space. Sky directions provide only two coordinates. A third requires an estimate of distance.
Large surveys use redshift, the displacement of spectral lines associated, among other things, with cosmic expansion. Together with an expansion model, it helps turn directions and spectra into a three-dimensional map. Galaxies’ own motions also affect redshift, so it is not a perfect tape measure; apparent cluster elongations along the line of sight require particular care. CfA: why galaxy catalogs need redshifts, Hogg: cosmological redshift and peculiar velocities.
Both panels below are invented. The sketch separates sample points from lines used to explain their pattern. A real research algorithm must establish connections using positions, uncertainties and classification rules instead of simply drawing attractive threads.
Hypothetical galaxies
The same points with teaching connections
- Each point is a hypothetical galaxy, not a star.
- Dense groups of points mark nodes.
- Lines highlight possible filaments; a pale gap represents an underpopulated volume.
Both panels are a teaching sketch with invented coordinates, not survey data or a physical simulation. Lines are drawn to explain the idea: they are neither luminous cables nor a measured dark-matter map. There is no physical scale; only part of a hypothetical network is shown.
A map also has limits. If a survey misses part of the sky or cannot detect faint galaxies, that region may contain few points even without a real void. Its volume boundaries, sensitivity and object selection therefore matter. A thick slice can superimpose distinct structures; a thin one may reveal only fragments of the network.
Observable matter also exists between galaxies
Diffuse gas often emits too faintly for an ordinary picture. But it can leave absorption lines in the light of a distant quasar, a bright active galactic nucleus. The background source illuminates matter along the path to us, and its spectrum carries that matter’s signature. Many such sight lines help connect gas with large-scale structure. NASA: studying intergalactic gas.
Under favorable conditions, gas-filament emission is detected too. In 2019, Umehata and colleagues described hydrogen Lyman-alpha observations of the SSA22 protocluster at a redshift of about 3.1. Detected structures extended more than one physical megaparsec in projection. Ionizing radiation from active galaxies in that unusual environment likely enhanced the emission. Original study: SSA22 gas filaments.
This is gas observed in a particular region, not a photograph of all the Universe’s dark matter. The authors did not even describe the proposed gas inflow into galaxies as directly measured in these data. The distribution of sources and emission supports a physical interpretation, but does not turn an image into a movie of the flow.
A void means lower population, not an absence of everything
One galaxy in a region does not automatically disqualify it as a void. Contrast with the mean environment on a specified scale matters. The dwarf galaxy KK 246, for example, lies in the Local Void. Other galaxies appear in its image, but some are far beyond that volume. NASA/ESA: a galaxy inside a cosmic void.
Branch · With numbersOptional · you can skip this
How a negative density contrast can describe a positive galaxy count
For galaxies in a selected sample, define number density n_(g) = N/V, where N is the count in volume V. Compare it with the mean \bar n_g using the contrast
\delta_g=\frac{n_g-\bar n_g}{\bar n_g}.
In hypothetical equal volumes, the mean corresponds to 40 galaxies, while an underdense region contains 8. Then δ_(g) = (8 − 40)/40 = − 0.8. Galaxies are present; their density is 20% of the mean. The deviation is negative, not the mass or the count itself. Eighty galaxies in that same volume would give δ_(g) = 1.
These are teaching numbers, not a catalog. We count galaxies selected in one way, not all mass: galaxies do not perfectly trace dark matter. The result depends on scale, boundaries and sample completeness; one equation does not provide a universal void-classification rule.
First establish what the picture represents
A web image may be a galaxy catalog, a mass reconstruction from gravitational lensing, or a computer simulation. Colors can mean different things in each case. Lensing uses distortions of distant sources’ images by foreground gravity; its mass map is inferred from measurements and a model. A simulation instead computes matter’s evolution from specified initial conditions. NASA: different ways of mapping the web.
Check your interpretation: if a purple line connects two galaxies on a map, does it prove that dark matter was photographed? No. Read the legend: it may represent a model connection, a false-color mass estimate or measured gas emission—different results with different limits.
The cosmic web explains how galaxies belong to a larger structure that gravity grew from early irregularities. To read its map, look for three things: what a point or color represents, what volume and selection are included, and where observation ends and reconstruction begins.