Nicole McPherson's team at Adelaide did not send anything to space. They loaded human, mouse and pig sperm into a microchannel slide, plugged both ends, and set it inside a box that rotates on two independent axes according to a random walk algorithm. The box lives in an ordinary laboratory incubator at 37 degrees. It brings the sample below 0.1g within two minutes.

Which is to say it does not remove gravity. Gravity is still there, pulling on everything in the room, including the box. What the box does is change gravity's direction faster than a cell can average it out. The cell is not weightless. The cell is confused.

Everything written this week about human reproduction beyond Earth rests on the proposition that those two states are close enough.

The best section of this paper is the one establishing that the box is not simply stirring the sample. A rotating vessel mixes fluid. If it mixes fluid, then sperm arriving at the far end of a channel have not necessarily swum there, and the entire experiment collapses into a very expensive demonstration of convection.

So the team ran the controls. They tested how sample position relative to the rotational axis affected motility and found that outer channels degraded it, probably through fluid shear, so they used only the center wells for everything afterward. They put phenol stained media and fluorescent beads in the channel openings to confirm that nothing was being carried across passively. They washed the seminal plasma out of every human sample because viscosity differences were causing mixing. They ran rotational controls, spinning the samples without nullifying gravity, to separate the effect of the machine from the effect of the simulated condition.

The rotational control caught something. Curvilinear velocity dropped 20 percent in the rotational control (P = 0.04), a decline that did not appear in the simulated microgravity condition. The control performed worse than the treatment on one measure. They reported it anyway.

None of this made the coverage. It is the reason to believe the coverage.

Human sperm under simulated microgravity were significantly less likely to traverse the channel and reach the collection point, in standard media (P = 0.002) and in media formulated to mimic female reproductive tract fluid (P = 0.007). Total motility, progressive motility, curvilinear velocity and lateral head displacement were all unchanged. The sperm were swimming exactly as well. They were arriving less often. That distinction is the whole finding, and it points at orientation rather than propulsion.

In mice, four hours of co-incubation under simulated microgravity cut fertilization by about 30 percent as measured by two cell development (P = 0.03). Extend the window to 24 hours and fertilization rates recover to baseline, but the embryos pay for it later, with delayed cleavage and reduced blastocyst cell numbers (P = 0.04 for epiblast count). In pigs, both fertilization (P = 0.01) and expanded blastocyst formation (P = 0.005) fell.

Then the interesting part. Sperm that did complete the channel under simulated microgravity bound hyaluronan better than controls (P = 0.04). Hyaluronan binding correlates with lower DNA fragmentation and better ICSI outcomes. The embryos they produced had more epiblast cells. Making the journey harder did not just reduce the number of arrivals. It changed who arrived.

Buried in the methods, after the description of the random walk algorithm, is this: the gravity bias system is proprietary to Firefly Biotech, and the authors acknowledge that reliance on a proprietary device may present a limitation in reproducibility.

That is an unusually clean admission. The instrument that defines the independent variable is a commercial black box. Another lab can buy the same box, but cannot inspect what it does, and cannot build one. The paper's central claim is that a machine successfully impersonated space, and the machine's impersonation technique is a trade secret.

They published the control algorithms anyway, deposited to Zenodo, which is more than the situation required of them.

There is also the matter of what the box cannot contain. There is no radiation in an incubator in Adelaide. The paper is explicit about this being the point, since real spaceflight cannot separate microgravity from cosmic rays, and the clinostat can. The coverage then places these results a few paragraphs upstream of a section on galactic cosmic rays damaging embryonic DNA, and lets the reader do the stitching. Two different experiments, two different variables, one continuous narrative.