Aerosol filter sampling was conducted during the summers of 1993 and 1994 at the solar powered remote atmospheric sampling site located 28 km southwest of the GISP2 camp (72.58N, 38.46W, 3205 m) in 1993 and 10 km southwest of the GISP2 site in 1994. Particles were collected on 90 mm Teflon Zefluor filters (1 mm pore size). The filter holders were open faced and covered with a large polyethylene bucket to prevent the collection of snow on the filter. Vacuum pumps were located in a trench beneath the surface. Airflow was measured using a Rockwell gas meter. The filters were run for approximately 16 hours each day during routine sampling periods excluding episodes of fog. During intensive sampling periods filters were changed every eight hours. In all cases, the flow rate was approximately 85 l min-1. All handling of the filters was done by personnel wearing clean room garb and shoulder length plastic gloves. Filters were removed from the filter holders in the field and stored in airtight Pyrex bottles. All filters were kept frozen until analysis at the Laboratoire de Glaciologie et Geophysique de lEnvironement (LGGE) in Grenoble, France. In order to minimize the evaporation of volatile aerosols on the filter, extraction took place in a clean hood in a cold room at -12o C. First, 1.0 ml of ultra pure methanol was added to the bottle to wet the filter. Then 9.0 ml of deionized water were added to the bottle to dissolve the soluble particles. Secondary extractions were run on 5 of the 155 filters from the 1994 field season. The average efficiency of the primary extraction was determined to be 93% based on the amount of soluble ions in the secondary extraction. The filter extracts were analyzed by ion chromatography for fluoride, lactate, acetate, glycolate, formate, methane sulfonate, chloride, nitrate, sulfate, oxalate, sodium, potassium, ammonium, magnesium, and calcium by ion chromatography. Detection limits for all three chemical species were calculated as 3 times the standard deviation of the amount of material on the field blanks divided by the volume of air drawn through each filter. While each filter sample had its own detection limits based on the filtered air volume, the mean aerosol concentration detection limits for 1993 and 1994 are summarized in the detection limit table. The atmospheric concentrations were calculated by subtracting the mean chemical mass on the blank filters from the mass on the sample filter.