2003AJ....126.1849B

Revised (B-V)₀-Mg₂data, which are used to test reddening predictions, are presented for 402 elliptical galaxies. These reddening predictions can tell us both what the intrinsic errors are in this relationship among gE galaxy stellar populations as well as details of nearby structure in the interstellar medium (ISM) of our Galaxy, and of the intrinsic errors in reddening predictions. Using least-squares fits, the explicit 1 σ errors in reddenings predicted by the Burstein-Heiles (BH) method and the Schlegel and coworkers (IR) method are calculated, as well as the 1 σ observational error in the (B-V)₀-Mg₂for gE galaxies. It is found that in directions with E(B-V)<0.100 mag (where most of these galaxies lie), 1 σ errors in the IR reddening predictions are 0.006-0.009 mag in E(B-V), those for BH reddening predictions are 0.011 mag, and the 1 σ agreement between the two reddening predictions is 0.007 mag. The IR predictions have an accuracy of 0.010-0.011 mag in directions with E(B-V)≥0.100 mag, significantly better than those of the BH predictions (0.024-0.025). Both methods yield good evidence that gas-to-dust variations that vary by a factor of 3, both high and low, exist along many lines of sight in our Galaxy. Both methods also predict many directions with E(B-V)<0.015 mag, despite the difference in zero point that each has assumed. The ∼0.02 higher reddening zero point in E(B-V) previously determined by Schlegel and coworkers is confirmed, primarily at the Galactic poles. Independent evidence of reddening at the north Galactic pole (NGP) is reviewed, with the conclusion that directions still exist at the NGP that have E(B-V)≪0.01. Two lines of evidence suggest that IR reddenings are overpredicted in directions with high gas-to-dust ratios. As high gas-to-dust directions in the ISM also include the Galactic poles, this overprediction is the likely cause of the E(B-V)∼0.02 mag larger IR reddening zero point relative to that of BH.