Two tracers carry the whole argument. Molecular hydrogen is the most abundant molecule in a protoplanetary disk, and ionized neon marks the jets and the hot atomic outflow [4]. JWST's sensitivity is what lets the team pull a broad molecular wind apart from the neon-marked jet in the same object [5]. Without that separation there is no ordering, only a blend.
The mechanism underneath the ordering is more interesting than the ordering itself. In 2020, Ilaria Pascucci's group predicted molecular winds that were massive enough to absorb the star's X-ray photons at early ages, a prediction made without any direct view of molecular hydrogen [8]. Tracing that hydrogen directly, the new survey says the prediction holds [9]. So the early wind is also the shield: while accretion is still strong and the molecular component is thick, the high-energy radiation that would later heat gas past escape velocity cannot reach the disk. Photoevaporation does not switch on because the star changes. It waits for the earlier wind to thin out [7].
The mass at stake is easy to understate. A young disk carries roughly 100 times more gas than dust [12], which puts gas at about 99 percent of the disk's mass [13]. Dispersal is not the loss of one ingredient out of several; it is the loss of nearly the entire object, and the giant planets are the only things that need the missing 99 percent [11]. Set the few-million-year disk phase against the Sun's present 4.5 billion years and the window in which Jupiter could have been built is about one part in a thousand of the system's life so far [14].
Now the caveat that the framing tends to swallow. Uma Gorti's point that disk dispersal sets a clock for planet formation is a statement about physics [10], and Bajaj's "race against time" [16] is a good description of the situation. But the 72 systems were each observed once, drawn from the JWST archive, and assembled into a sequence because each one represents a different stage [1][3]. The stages are described by accretion activity and evolutionary stage, not by an age in years at which winds turn atomic [17]. What the survey delivers is an order of operations, with the conclusion that no single process runs the dispersal [15]. That is a real result, and it is ordinal. A deadline for gas-giant assembly is a different measurement, and this one does not make it.
