Evidence that microbubbles are the main sources of ultrasound contrast in injected solutions has been largely indirect. To investigate this directly, we examined freshly agitated indocyanine green, freshly agitated water, commercially prepared precision microbubbles (diameter 75 +/- 25 mu) in gelatin, carbonated water, "degassed" indocyanine green solution, and "degassed" water in one or more of four different assay systems. Only fluids with microbubbles produced ultrasound contrast. Injected contrast material rose in a water bath at a rate that identified it as being caused by microbubbles. Indocyanine green and gelatin surface tensions were measured and found to be low (43 dynes/cm2), thus explaining their tendency to stabilize the microbubbles that cause ultrasound contrast effect when injected and to hold foam after agitation. The force of hand injections (force similar to that used clinically through catheters and 19-gauge or 23-gauge needles) was below the force needed to cause cavitation or ultrasound contrast effect. Microbubble content could be quantified by the decrease in amplitude of the echo from a structure distant to the microbubbles. We conclude that that the ultrasound contrast effect seen in peripherally injected fluids is caused by microbubbles present in the injectant. The contrast is not due to cavitation at needle tips, and it can be quantified over a limited range. Improved design for a peripheral contrast agent is suggest.