Three baryonic systematics, each sufficient: on the reported redshift evolution of the MOND acceleration scale

Context. Ciocan et al. [2026a] report the first measurement of a redshift evolution of the characteristic acceleration scale a0 of the radial acceleration relation (RAR), from 79 star-forming galaxies at 0.33 < z < 1.44 in the MUSE Hubble Ultra Deep Field, obtaining a1 =1.59+0.11−0.10 × 10−10 m s−2(95% CI) and describing the detection as ∼30σ.Aims. I examine three systematics of the baryonic mass budget using only the authors’ own public catalogue. Each has a different character: one biases the trend, one is the trend restated, and one destroys the quoted precision.Results. (I) Unmodelled molecular gas. The model contains a stellar disk and atomic gas only. Evaluating Tacconi et al. [2018] per galaxy on the catalogue z, M⋆ and SFR, the missing mass rises from a median MH2 /M⋆ = 0.51 to 0.97 across the four redshift quartiles. Propagating it removes 82% of the binned trend and 77% of the fractional evolution, and the conclusion requires only that molecular gas evolve faster than (1 + z)1.06. (II) A drift in dynamical stellar masses. The dynamically inferred M⋆ falls relative to the photometric one by 0.39 dex between the extreme redshift bins (3.6σ), while the photometric mass is flat with z. The drift appears in both halo models that report a stellar mass, the underlying Mvir decline in all six, and it is opposite in sign to the expected stellar-to-halo-mass evolution. Its size matches the +0.2 to +0.45 dex that Ciocan et al. state would be required to remove the trend. (III) Disk–halo degeneracy. The same galaxy is assigned halo masses differing by a median 1.04 dex across the six released halo models.Conclusions. Each systematic is individually of the size of the reported signal, and all three live in the baryonic mass budget rather than in the kinematics. a1 should not presently be used to constrain modified-gravity models.