Evidence for Evolution or Bias in Host Extinctions of Type 1a Supernovae at High Redshift

Type 1a supernova magnitudes conventionally include an additive parameter called the extinction coefficient. We find that the extinction coefficients of a popular ``gold'' set are well correlated with the deviation of magnitudes from Hubble diagrams. If the effect is due to bias, extinctions have been overestimated, which makes supernovas appear more dim. The statistical significance of the extinction-acceleration correlation has a random chance probability of less than one in a million. The hypothesis that extinction coefficients should be corrected empirically provides greatly improved fits to both accelerating and non-accelerating models, with the independent feature of eliminating any significant correlation of residuals.