&NA;Real time intraoperative sonography is a valuable tool for visualizing subcortical mass lesions. Although most solid lesions are hyperechogenic, little is known about the ultrasound characteristics of intracerebral hemorrhage and subsequent hematoma formation. We sought to determine the sonographic characteristics of an acute intracerebral hematoma, to study its evolution, and to explore factors responsible for its echogenicity. Acute intracerebral hematomas were created in adult mongrel dogs using heparinized or unheparinized autologous whole blood injected under sonographic visualization into the centrum semiovale of 10 cerebral hemispheres. Different components of blood were also imaged in polyurethane test tubes (n = 56) immersed in a degassed. room temperature water bath. All studies were performed with continuous ultrasound recording using the ATL real time Neurosector scanner with the variable 3‐, 5‐, and 7.5‐MHz transducer. Intracerebral hematomas were initially hypoechogenic, becoming hyperechogenic between 16 and 23 seconds after injection (average, 22 seconds). Full echogenicity was obtained between 40 and 213 seconds after injection (average. 91 seconds). There was no difference between the sonographic appearances of hematomas formed with whole blood and those formed with heparinized whole blood. In addition, we demonstrated the superior sensitivity of the 7.5‐MHz frequency in visualizing intracerebral hematomas in vivo, as opposed to the 3‐ and 5‐MHz frequencies. Whole blood, heparinized whole blood, and citrated whole blood were found to be highly echogenic in vitro. Phosphate‐buffered saline, plasma, serum, and packed red blood cells (PRBCs) were hypoechogenic. Resuspended PRCBs in phosphate‐buffered saline, plasma, or serum were echogenic. Our data suggest that hyperechogenicity is independent of the clotting mechanism and is related to blood stasis. The primary factor in stasis accounting for blood echogenicity seems to be red cell aggregation. We conclude that: (a) acute intracerebral hematomas are initially hypoechogenic, becoming hyperechogenic by 22 seconds after their formation: (b) all acute hematomas are hyperechogenic within 4 to 5 minutes; (c) echogenicity seems to be secondary to red cell aggregation and independent of the clotting mechanism; (d) good image resolution of acute hematoma nessesitates relatively high frequency sonographic imaging because of the small size of the echo‐producing red cell aggregates. (Neurosurgery14:48‐51, 1984)