Directional Anisotropy Gravity — Predictions, SPARC Verification, and Operational Tests

A theory is only as useful to observers as the predictions it puts within reach of an instrument. This companion to Directional Anisotropy Gravity (DOI 10. 5281/zenodo. 20397188) is written for that purpose: to make it possible to plan, execute, and interpret a test of the demonstrated core of DAG without reading the full substrate construction. The document is structured around the three co-equal primary predictions the demonstrated core commits to on galactic scales: — Primary I: history-dependent residual gravity at fixed present baryons. The galactic dark-matter phenomenology is the inhomogeneous projection of a single memory field; the source that determines the residual gravity is the history-weighted effective mass Mbᵉff, not the instantaneous present-day baryon mass Mb (t₀). — Primary II: a universal activation-radius ratio rₜ / rAQUAL = ε⁻¹ᐟ² ≈ 7–10, independent of galactic mass, where ε is the small dimensionless ratio fixed by the substrate. The SPARC verification on 149 quality-gated galaxies is reported in full as the first empirical test of this prediction: the measured mean sits at the upper edge of the predicted band, and the Spearman correlation with baryon mass is consistent with zero, ruling out the naive instantaneous-baryon reading at high significance. — Primary III: non-spherical memory multipoles with a logarithmic-monopole far field and angular decay r^ (-√ (ℓ (ℓ+1) ) ). The transport equation predicts a definite multipolar far-field structure that current and near-future kinematic and lensing surveys can target. The bridge lemma rₜ / rAQUAL (t₀) = ε⁻¹ᐟ² · (Mbᵉff / Mb (t₀) ) ^ (1/2) ties Primaries I and II into a single consistency cycle: a measured deviation from the universal band at fixed Primary II must come from a history-weighting deviation in Primary I, and vice versa. A set of structural null boundaries is included as falsifiers in the strict sense: — weak equivalence principle: δa / a ≲ 10⁻²⁶ derived, with any composition-dependent violation at the 10⁻¹⁵ MICROSCOPE level ruling out the demonstrated core;— solar-system PPN: βPPN = γPPN = 1 derived exactly, with any deviation at the LLR / Cassini level a falsifier;— photon sector: no leading non-minimal RF² coupling at any sub-Planck scale;— Standard-Model attachment: exactly three light chiral families and no light exotics. For each test the document specifies the data requirements, the dominant nuisance parameters, and the quantitative thresholds at which the test yields support, tension, or falsification. The minimum prior reading is the abstract and Section 1 of the core paper; the operational equations are reproduced in full here. The accompanying SparcTest. zip archive contains the SPARC analysis code and gated-sample tables used in the Primary II verification reported in this document. Companion documents in the DAG public set: — Core paper (DOI 10. 5281/zenodo. 20397188): the theory and its full justification. — Technical Supplement (DOI 10. 5281/zenodo. 20397317): full proofs, assumption ledger, and dependency tree. — Programme sourcebook v12. 2 (DOI 10. 5281/zenodo. 20071356): the full research programme; open quantitative items and longer-horizon predictions live there.