Coefficient of Haze (CoH) was the official proxy for the mass concentration of particulate matter in the US from ca. 1950 onwards. Originally, a linearity between CoH and sample load was claimed, but although shortly after the introduction of the method it was found that the relation between mass loading and the fraction of light 10 absorbed by the sample was identical to the nonlinear curve of the European "Black Smoke" (BS) method this correction was not officially adopted. We found that BS is a linear proxy for Elemental Carbon (EC) enabling an easy nonlinear conversion of CoH to EC concentration. Earlier it had been assumed that light-absorption by a sample is equal in the two methods, although CoH is obtained by light transmission while BS by light reflection. The "Filter Smoke Number" (FSN), used for diesel-soot emissions, is in essence the light-absorption by samples collected on the same filter type as in the CoH method (but probed in reflection). We noticed that the curve relating load and FSN is indistinguishable from the BS/EC curve. Ingram showed that light-absorption measured in reflection and transmission for this filter type is highly similar, so the standard BS/EC curve can be used to compare BS/EC and CoH. At low loadings the relation of EC concentration and CoH was quite similar to the equivalency factor of Black Carbon (BC) concentration and CoH. At CoH = 8, the historic limit of a severe smog day; however, the EC/BC-concentration alone was three times higher than that based on a (linear) equivalency factor. At CoH = 8, EC/BC mass concentrations alone reached the OSHA 8-hr workplace exposure limit of 150 µg m −3 for total carbon (the sum of organic carbon (OC) and EC).Implications: In this MS we show that the historic database of Coefficient of Haze (CoH) data can be converted to Elemental Carbon (EC) concentrations via several steps involving other historic and current measurement techniques (the European Black Smoke (BS) technique and techniques to measure Black Carbon (BC) and EC). The originally claimed linear relation between CoH and sample load is in reality a strongly nonlinear relation, so using the original linear relation leads to severe underestimations of historic EC concentrations especially on days with high concentrations of EC.