A reconnaissance investigation of hydrology and water quality was conducted to evaluate possible sources of bacteria and nutrient contamination in the Cave Springs Branch basin and the underlying karstic Ozark Plateau aquifer system. Objectives were to: (1) determine the directions of ground-water flow in the basin and determine whether Cave Springs Branch interacts with ground water, (2) compare water quality in Cave Springs Branch with water quality in nearby wells to determine whether the stream is contaminating nearby wells, and (3) determine sources of fecal coliform bacteria and nitrate contamination in Cave Springs Branch and ground water. Potential sources of bacteria and nitrate in the area include cultivated agriculture, cow and horse on pasture, poultry production, households, and wildlife. Presence of fecal coliform and fecal streptococcal bacteria directly indicate fecal contamination and the potential for the presence of other pathogenic organisms in a water supply. Nitrate in drinking water poses health risks and may indicate the presence of additional contaminants. Fecal coliform bacteria colony counts were least in wells, intermediate in the poultry-processing plant wastewater outfall and Honey Creek above the confluence with Cave Springs Branch, and greatest in Cave Springs Branch. Bacteria strains and resistance to antibiotics by some bacteria indicate that livestock may have been sources of some bacteria in the water samples. Multiple antibiotic resistances were not present in the isolates from the water samples, indicating that the bacteria may not be from human or poultry sources. Ribotyping indicates that Escherichia coli bacteria in water samples from the basin were from bird, cow, horse, dog, deer, and human sources. The presence of multiple ribotypes from each type of animal source except bird indicates that most of the bacteria are from multiple populations of source animals. Identifiable sources of bacteria in Cave Springs Branch at the state line were dominantly cow and horse with one ribotype from bird. Escherichia coli was detected in only one well sample. Bacterial ribotypes in water from that upgradient well indicated human and dog feces as sources for bacteria, and that on site wastewater treatment may not always be adequate in these highly permeable soils. Greater concentrations of nitrate in Cave Springs Branch and O'Brien Spring relative to the poultry-processing plant wastewater outfall may be due, in part, to conversion of ammonia from poultry processing plant wastewater. The poultry-processing plant wastewater outfall sample collected in March 2000 contained greater concentrations of ammonia and total organic nitrogen plus ammonia than the spring, stream, and well samples collected during August 1999. Cave Springs Branch and Honey Creek contributed approximately equal loads of nitrogen to Honey Creek below the confluence and the greatest loads of nitrogen were introduced to Cave Springs Branch by the poultry processing plant wastewater outfall and O'Brien Spring. Nitrate concentrations in upgradient well samples ranged from 0.38 to 4.60 milligrams per liter, indicating that there are sources of ground-water nitrogen other than Cave Springs Branch, such as animal waste, fertilizer, or human waste. Nitrogen compounds in water from wells downgradient of Cave Springs Branch may be from Cave Springs Branch, fertilizers, animal waste, or human waste.