The CMB comes from a last-scattering surface: along each direction, we receive photons that stopped being repeatedly scattered when the early universe became transparent. It is an electromagnetic horizon, not the outer edge of all that can exist or be observed by every possible messenger.

The emission occurred about 380,000 years into the hot Big Bang chronology. Electrons and nuclei combined into neutral atoms, sharply increasing the distance photons could travel. Expansion has since stretched those photons to microwave wavelengths. We see a young cosmos across the whole sky, not a shell located at one permanent distance from every observer.

In 1964, Arno Penzias and Robert Wilson at Bell Labs found an excess antenna temperature that remained in every direction. They checked the instrument and removed pigeon nesting material, but the signal persisted. A Princeton group led by Robert Dicke was preparing to search for the relic radiation predicted by hot-universe models; contact between the teams supplied the interpretation. The paired papers appeared in 1965.

Penzias and Wilson shared the 1978 Nobel Prize for the discovery. The CMB became decisive evidence that the observable universe passed through a much hotter, denser phase. It did not demonstrate that this phase was the absolute beginning of space, time or matter.

Precision satellite missions turned the already detected CMB into a precision map. COBE (1989-1993) measured its near-perfect blackbody spectrum and first detected large-scale anisotropy.

WMAP (2001-2010) mapped the full sky with much finer resolution. The temperature variations are typically only tens of microkelvin, roughly parts in 100,000. They trace a mixture of density, velocity and gravitational effects near last scattering.

The Planck mission (2009-2013) mapped temperature and polarization at still finer angular scales. Within a cosmological model, those measurements constrain the universe's age, geometry and contents.

The ripples are not a complete blueprint of every later object. They are early conditions from which gravity, gas physics and many contingent histories helped form the cosmic web, galaxies, stars and planets.

Inflation is a leading family of models for a very early period of accelerated expansion. It offers accounts of the universe's large-scale uniformity and of nearly scale-invariant primordial fluctuations that later helped structure grow. CMB temperature and polarization data strongly constrain inflationary models and are consistent with several of them.

Consistency is not a unique detection. Other early-universe scenarios can reproduce some of the same observables, and the physical origin and detailed mechanism of inflation remain unsettled. The CMB lets theory reach toward energies no accelerator can reproduce, but the reach is made through model-dependent inference.

Within the standard flat ΛCDM model, Planck CMB data combined with other observations constrain a present-day cosmic budget of roughly five percent baryonic matter, one quarter dark matter and about two thirds dark energy. These are model-dependent parameter estimates, not three substances weighed directly by the map.

Dark matter is inferred through gravitational effects; whether it has other non-gravitational interactions is unknown. "Dark energy" names whatever accounts for the observed accelerated expansion within the model, often represented by a cosmological constant rather than a force. CMB structure also places tight limits on large-scale spatial curvature when combined with other data.

The CMB links observation to an early chapter of cosmic history. Its small variations trace density, motion and gravitational potential near last scattering. Under gravitational evolution, those initial variations are connected to the later cosmic web.

This is the moment the universe became readable in light. Earlier epochs are approached through the imprints they left, not seen directly with photons. The distinction matters: the map is an extraordinary record of cooling and expansion, while questions about an absolute beginning remain beyond what the CMB alone can answer.

The experimental roadmap is still changing. US agency support for CMB-S4 ended in July 2025; the 2026 Astronomy and Astrophysics Advisory Committee report notes that existing and upgraded instruments can pursue parts of its science, though not at the sensitivity the full project was designed to reach. JAXA's LiteBIRD passed a second mission-definition review in June 2026 and entered project preparation, aiming for launch in Japanese fiscal year 2036. It will search all-sky polarization for primordial B modes: a demanding test of some inflationary models, not a proof of an absolute origin.