Same as Fig. 16, but now for a comparison of model s12A (M(⁵⁶Ni) = 0.05M_⊙) and s15p2 (M(⁵⁶Ni) = 0.063M_⊙) to SN 1987A (M(⁵⁶Ni) = 0.069M_⊙). Both models adopt a solar metallicity, not exactly consistent with the LMC metallicity for SN 1987A. The lower-mass progenitor does a good job at matching the three strongest lines, but severely underestimates the Ca II triplet, and does a poor job with at matching the Fe emission forest shortward of 6000 Å. By contrast, the higher mass model does a better job with the Fe emission, overestimates the O I doublet, and underestimates the Ca II line strengths. The flux normalization is done at 6800 Å.