A third of the universe's ordinary matter was unlocatable for twenty years, until localized fast radio bursts closed the census: the Macquart dispersion-redshift relation accounts for the full BBN-CMB baryon budget, with about 80 percent in the diffuse intergalactic phase and eROSITA now imaging WHIM filaments directly. The books balance, but the model that misplaced its visible sector for decades must now explain the measured partition, which its feedback prescriptions span by tens of percent.
ΛCDM distributes baryons as passengers on a dark matter web, with their final whereabouts set by feedback physics the model cannot derive. The baryon locations are outputs of its least constrained ingredient, so the model could neither find the gas in advance nor predict its partition now that FRBs measure it.
SCT replaces the dark matter scaffold with collision-deposited structure: the cosmic web is where the cascade put the baryons, filaments from head-on collision geometries and walls from compression shells (P33, P34), and there is no second invisible matter sector. That raises the stakes on the budget: with no dark matter particle to absorb discrepancies, the baryon census must close exactly, and the cluster-scale identity A* × f_b = f_b,cosmic, confirmed at 0.1675 across the X-COP sample, already requires every baryon to be present and accounted for in bound systems. The FRB closure of the global budget is the diffuse-phase half of the same requirement.
The measured partition is the deposition map. SCT expects the dominant baryon share strung along the collision-built filament network as warm-hot gas, because the cascade thermalized matter where it deposited it (P22, P25), with a relic entropy floor rather than a feedback-sculpted distribution; eROSITA's stacked filament WHIM is that prediction made visible. The convergent mesh sum (P62) keeps the bound-system bookkeeping insensitive to the diffuse tail, which is why the cluster identity held before the census closed.
This is the same deposition geometry behind the void emptiness and the giant structures: one map, several censuses. There is no need to invoke tuned feedback violence to put the gas where it is observed.
The closing census is the test. FRB samples of thousands measuring a cosmic baryon density inconsistent with the BBN-CMB value, or a partition with the diffuse share well below the filament-dominated expectation, would break the deposition map. At bound scales, the registered kill stands: the product A_obs × f_b varying by more than a factor of two across twenty or more relaxed clusters falsifies the identity that obliges the full baryon budget to be present.