Transport and collection characteristics were studied for gaseous elemental mercury (Hg(0)(g)) in natural gases using newly developed methodology based on amalgamation, isotope dilution with permeation tubes and inductively coupled plasma mass spectrometry. The study involved different Au-Pt collection tube designs, tubing materials and gaseous matrices, including air, natural and sales gas, as well as methane and sales gas to which hydrogen sulfide (H(2)S) had been added. The Hg(0)(g) capacity of the Au-Pt tubes was determined to 3.5 +/- 0.1 microg. Blanks and detection limits of gaseous mercury (Hg(g)) were 58 +/- 17 pg m(-3) and 50 pg m(-3), respectively, for a 60 L sample volume. For the gases tested, added Hg(0)(g) tracers could be collected with 90% or higher efficiency at flow rates and volumes of up to 10 L min(-1) and 100 L, respectively. The collection efficiency was found to be independent of the type of gas tested, even in the presence of H(2)S. However, for the gases containing H(2)S, the apparent transport efficiency of added Hg(0)(g) tracers through stainless steel tubing varied from 50 to 150% upon changing the temperature from 25 to 100 degrees C. The interaction of stainless steel with Hg(0)(g) leading to either a sink, or source of Hg, was not observed in the absence of H(2)S, nor was it observed for PTFE tubing in the presence of H(2)S. These observations raise questions about the applicability of currently used sampling procedures for determination of Hg(g) in H(2)S rich natural gases, including the 6978-2 ISO standard method, in which stainless steel is a prescribed material for tubing and valves of the sampling apparatus.