Detection capabilities: some historical footnotes

Part I Summary of relevant topics from 1923 to present—including: Currie (Anal Chem 40:586–593, 1968 ) detection concepts & capabilities; International detection & uncertainty standards; Failure of classical 14 C dating and birth of new scientific disciplines; Exploratory nuclear data analysis of 85 Kr monitors found coincident with the collapse of the Iron Curtain (1989); Faulty statistics proved responsible for mistaken assertions that Currie’s L C yields excessive false positives; Low-level counting and AMS for atmospheric 37 Ar and µmolar fossil/biomass carbon in the environment; Erroneous assumption that our low-level background is a Poisson Process, linked to ~8 % spurious anticoincidence events. Part II Exact treatment of bivariate Poisson data—solved in 1930s by Przyborowski and Wilenski, Krakow University, for detecting extreme trace amounts of a malicious contaminant (dodder) in high purity seed standards. We adapted their treatment to detection capabilities in ultra-low-level nuclear counting. The timing of their work had great historical significance, marking the start of World War II, with the invasion of Poland (1939).