The purpose of this study was to determine and compare the accuracy of palpation per rectum and linear-array ultrasonography for diagnosing follicular vs luteal ovarian cysts in cows. Forty-seven examinations of ovarian cysts from 28 cows were diagnosed by palpation per rectum as either a firm, thick-walled structure (luteal cyst) or a soft, thin-walled structure (follicular cyst) during weekly herd examinations. The ovaries of each cow were then examined by ultrasonography. Ultrasonograms of cysts greater than 25 mm in diameter were diagnosed as luteal or follicular cysts and were recorded on videotape for evaluation by a second clinician. Serum progesterone concentrations at the time of examination were determined by radioimmunoassay and used to classify luteal (greater than 0.5 ng/ml) or follicular (less than or equal to 0.5 ng/ml) cysts. Selection of this discriminatory level was based on response of a proportion of cows with luteal cysts that were given 25 mg of prostaglandin F2 alpha at the time of diagnosis by ultrasonography. Sensitivity and specificity of palpation per rectum for diagnosis of type of ovarian cyst were low (43.3 and 64.7%, respectively). In contrast, sensitivity and specificity of ultrasonography were considerably higher (86.7 and 82.3%, respectively). Agreement between the 2 methods of diagnosis was 57.4%. Overall agreement between the 2 clinicians' diagnoses by ultrasonography was 85.1%. On the basis of our findings, we confirm that luteal and follicular cysts cannot be accurately differentiated by palpation per rectum alone. These data suggest that linear-array ultrasonography is more effective than palpation per rectum for diagnosing type of ovarian cyst in cows.