Maher has described a procedure for the determination of total arsenic in marine sediments. Many factors explain the poor performance of the direct injection graphite furnace atomic absorption spectrometry method for seawater: barium carbide formation within the heated graphite tube during atomisation; furnace emission at the analytical wavelength; smoke from sodium chloride at atomisation; barium ionisation; and chloride vapour phase interferences. Hydride generation and atomic absorption detection provide a very sensitive method for metalloid analyses. Collection of the hydride in a liquid nitrogen trap and then subsequent rapid introduction into an atomiser improved the sensitivities of such elements as arsenic and tin by an order of magnitude compared to the continuous method. The microbial methylation of inorganic mercury to methyl mercury observed in lakewater sediments has also been observed in marine sediments. Mercury has been found in some sediments but at very low concentrations, mainly from areas of known mercury pollution.