Five insecticide residues and 20 herbicide residues were detected in water samples collected from 50 shallow wells screened in the surficial sand and gravel aquifer in Suffolk County, Long Island in areas with known or suspected residues. Laboratory analytical methods with extremely low detection limits - from 0.001 to 0.2 ?g/L (micrograms per liter) - were used to analyze the samples for 60 pesticide residues. Forty-four of the samples contained at least one pesticide residue, and some samples contained as many as 11 different pesticides or pesticide metabolites. Only four water- quality standards were exceeded in the samples collected in this study. Dieldrin exceeded the New York State Class GA standard (0.004 ?g/L) in samples from eight wells. The Federal and New York State Maximum Contaminant Level for simazine (4 ?g/L) was exceeded in samples from two wells, and the State Class GA standard for simazine (0.5 ?g/L) was exceeded in samples from six wells. Federal water-quality standards have not been established for many of the compounds detected in this study, including herbicide metabolites. Maximum concentrations of four herbicide metabolites -metolachlor ESA (ethanesulfonic acid), metolachlor OA (oxanilic acid), and the alachlor metabolites alachlor ESA and alachlor OA -exceeded 20 ?g/L. The maximum concentration of one herbicide (tebuthiuron) exceeded 10 ?g/L, and the maximum concentration of three herbicides (simazine, metolachlor, and atrazine) and one herbicide metabolite (deisopropylatrazine) ranged from 1 to 10 ?g/L. The herbicide metolachlor, which is used on potato fields in Suffolk County, and its metabolites (metolachlor ESA and metolachlor OA) were most frequently detected in samples from agricultural areas. The herbicides simazine and tebuthiuron, which were used in utility rights-of-way, and the simazine metabolite deisopropylatrazine were detected at concentrations greater than 0.05 ?g/L most frequently in samples from residential and mixed land-use areas. The results of this investigation are not necessarily representative of conditions throughout the remainder of Long Island, because these samples were collected in areas of known or suspected residues.